Search

Background: Automatic speech recognition (ASR) can potentially help older adults and people with disabilities reduce their dependence on others and increase their participation in society. However, maxillectomy patients with reduced speech intelligibility may encounter some problems using such technologies. Objectives: To investigate the accuracy of three commonly used ASR platforms when used by Japanese maxillectomy patients with and without their obturator placed. Methods: Speech samples were obtained from 29 maxillectomy patients with and without their obturator and 17 healthy volunteers. The samples were input into three speaker‐independent speech recognition platforms and the transcribed text was compared with the original text to calculate the syllable error rate (SER). All participants also completed a conventional speech intelligibility test to grade their speech using Taguchi's method. A comprehensive articulation assessment of patients without their obturator was also performed. Results: Significant differences in SER were observed between healthy and maxillectomy groups. Maxillectomy patients with an obturator showed a significant negative correlation between speech intelligibility scores and SER. However, for those without an obturator, no significant correlations were observed. Furthermore, for maxillectomy patients without an obturator, significant differences were found between syllables grouped by vowels. Syllables containing /i/, /u/ and /e/ exhibited higher error rates compared to those containing /a/ and /o/. Additionally, significant differences were observed when syllables were grouped by consonant place of articulation and manner of articulation. Conclusion: The three platforms performed well for healthy volunteers and maxillectomy patients with their obturator, but the SER for maxillectomy patients without their obturator was high, rendering the platforms unusable. System improvement is needed to increase accuracy for maxillectomy patients. [ABSTRACT FROM AUTHOR]

Data-driven platforms with rich data and learning analytics applications provide immense opportunities to support collaborative learning such as algorithmic group formation systems based on learning logs. However, teachers can still get overwhelmed since they have to manually set the parameters to create groups and it takes time to understand the meaning of each indicator. Therefore, it is imperative to explore predictive indicators for algorithmic group formation to release teachers from the dilemma with explainable group formation indicators and recommended settings based on group work purposes. Employing learning logs of group work from a reading-based university course, this study examines how learner indicators from different dimensions before the group work connect to the subsequent group work processes and consequences attributes through correlation analysis. Results find that the reading engagement and previous peer ratings can reveal individual achievement of the group work, and a homogeneous grouping strategy based on reading annotations and previous group work experience can predict desirable group performance for this learning context. In addition, it also proposes the potential of automatic group formation with recommended parameter settings that leverage the results of predictive indicators.

Individual identification plays a pivotal role in ecology and ethology, notably as a tool for complex social structures understanding. However, traditional identification methods often involve invasive physical tags and can prove both disruptive for animals and time-intensive for researchers. In recent years, the integration of deep learning in research has offered new methodological perspectives through the automatisation of complex tasks. Harnessing object detection and recognition technologies is increasingly used by researchers to achieve identification on video footage. This study represents a preliminary exploration into the development of a non-invasive tool for face detection and individual identification of Japanese macaques (Macaca fuscata) through deep learning. The ultimate goal of this research is, using identification done on the dataset, to automatically generate a social network representation of the studied population. The current main results are promising: (i) the creation of a Japanese macaques' face detector (Faster-RCNN model), reaching an accuracy of 82.2% and (ii) the creation of an individual recogniser for the Kōjima Island macaque population (YOLOv8n model), reaching an accuracy of 83%. We also created a Kōjima population social network by traditional methods, based on co-occurrences on videos. Thus, we provide a benchmark against which the automatically generated network will be assessed for reliability. These preliminary results are a testament to the potential of this approach to provide the scientific community with a tool for tracking individuals and social network studies in Japanese macaques., (© 2024. The Author(s), under exclusive licence to Japan Monkey Centre.)

This paper describes an automatic quiz generation system designed to support language learning that utilizes digital textbook logs. Learners often memorize words in digital textbooks while preparing for an examination, and they often use the highlight function for the words. Previous studies regarding annotations and highlights have shown that learning only by using the highlight function on important content in textbooks did not affect learning achievements. Therefore, in this study, we developed a system that can support the repeated learning by analyzing digital textbook logs and providing appropriate quizzes. An evaluation experiment involving 31 international students was conducted to assess whether the quizzes provided by our proposed system are able to enhance the learning achievements as compared to teacher-created quizzes. The results show that the quizzes by our proposed system and the teacher-created quizzes were both equally effective. A correlation analysis was conducted to identify the correlation among the learning achievements, the number of quizzes, and each variable in questionnaires. We found that there is a positive correlation between the number of quizzes and the students' learning achievements.

Recently, a retail industry has trended to adopt automatic checkout systems for enhancing customer sales. Typically, an automatic checkout system employs an object detection technique (i.e., image processing) based on a commonly used deep learning model (e.g., Faster R‐CNN and YOLO) that has been trained for items. Therefore, it takes much time to relearn the model whenever an item is added to the system. To save the relearning cost, we propose a novel automatic checkout system reducing the relearning cost of an additional item. © 2024 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Cirrhosis is a liver disease resulting from abnormal continuation of fibrosis, and ultrasound imaging is widely used for cirrhosis diagnosis because of its non‐invasiveness. However, due to unclear appearances of cirrhosis on ultrasound images, diagnoses are difficult and individual results possibly differ depending on the physician's experience. Recently, computer‐aided diagnostic systems using image processing and machine learning have been developed to help physicians detect cirrhosis as a 'Second opinion'. Some related studies have focused on a scenario where physicians set ROIs (Region of Interests) manually because selecting reliable ROIs for training a classifier and classification of patients is indispensable. But, the accuracy of such systems depends inherently on the quality of ROIs, and thus the workloads of physicians increase. In this paper, we propose a reliability evaluation method (REM) for each ROI based on its posterior probability and relationship to peripheral ROIs. The assumption of our proposal is that reliable regions of cirrhosis and normal can be observed in certain regions predominantly. We evaluated the effectiveness of the REM and its optimization for practical use. Experimental results showed that our proposed method curated reliable ROIs and improved classification performance in terms of AUC (Area Under the Curve). © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Automatic Singing Label Calibration (ASLC) aims to enhance the labeling accuracy of coarse singing labels through the analysis of raw audio. However, the ASLC model faces limitations due to the challenges and costs associated with generating or augmenting real‐world songs. To address this problem, we propose a novel approach to strengthen limited singing audio using easily available musical instrument audio. Directly using the musical instrument audio as a data augmentation for the singing audio is unreliable due to the distinct differences between vocal and instrumental sounds. Therefore, we employ transfer learning, which allows relevant knowledge to be transferred from one domain to another. In the pre‐training stage, the ASLC model learns to predict the accurate labels from the musical instrument audio. We then consider the vocal as a special musical instrument and fine‐tune the pretrained ASLC model using a singing annotation data set. Experimental results demonstrate that our transfer learning‐based approach outperforms the original ASLC model. By leveraging the readily available musical instrument audio, our method achieves improved performance in enhancing the labeling accuracy of singing audio. © 2024 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In this study, the viral genome extraction performance of automatic nucleic acid extractors and manual nucleic acid extraction kits was compared. We showed that compared with manual kits, the automatic extractors showed superior genome extraction performance using bovine viral diarrhea virus (BVDV) genome-positive cattle sera and bovine coronavirus/infectious bovine rhinotracheitis virus-spiked cattle nasal swabs. In addition, the subgenotyping of BVDV strains detected in Tokachi Province in Japan during 2016–2017 was performed. Results showed that most of these BVDV strains belonged to subgenotype 1b, while few strains belonged to subgenotypes 1a and 2a. This study showed the high applicability of automatic nucleic acid extractors in extracting multiple viral genomes and the dominant subgenotype of BVDV in Tokachi. [ABSTRACT FROM AUTHOR]

A 79-year-old woman collided with a cliff in a passenger automobile. The fire department acknowledged an automated collision notification from the D-Call Net (DCN) at 1 min after the accident and called for doctors by helicopter ( "Doctor-Heli" [DH] in Japan) 9 min after the injury. The DH reached the victim 28 min after the injury, and examination revealed pain in the right side of her chest, tachypnea, and a weak radial artery pulse (indicating shock). The DH arrived at the hospital 49 min after the injury. Thoracic drainage was performed for right-sided tension pneumothorax. She recovered from shock but was diagnosed with flail chest and placed on a respirator. She was extubated on postoperative day 6 and transferred to a rehabilitation hospital on postoperative day 57. Because of the DCN, the patient received treatment 15 min earlier than she would have with the conventional system. Emergency response task forces must develop strategies for connecting DCN warnings to rapid medical response systems.

We have explored tsunami current signals in maritime Automatic Identification System (AIS) data during the 2011 Tohoku, Japan, tsunami. The AIS data were investigated in detail taking into account ship motion and response to tsunami current. Ship velocity derived from AIS data was divided into two components in terms of the ship heading: heading-normal and heading-parallel directions. The heading-normal velocity showed good agreement with the simulated tsunami current, as mentioned in our former research. Here, we found the heading-normal velocity was contaminated by non-tsunami noises that were mostly related to the ship yaw motion around the pivot point. The noises due to the yaw motion were reasonably corrected in the heading-normal velocity. The corrected heading-normal velocity clearly showed better agreement with the simulated tsunami current. Although the heading-parallel velocity is basically the navigation speed, and is mostly controlled by ships' captain, we could find the heading-parallel velocity was also drifted by tsunami currents. The corrected heading-normal velocity was still a ship response to the tsunami current. Based on an equation of a ship response to tsunami currents, we numerically estimated tsunami current from the corrected heading-normal velocity. We could find very slight improvements in estimating the tsunami currents, which indicated that this operation possibly worked as a secondary correction. Tsunami currents of tens of centimeters per second are expected to be suitably detected using AIS based on discussion on detection limit. [ABSTRACT FROM AUTHOR]

This paper is an extension of research originally reported in [1]. Here, we propose a novel method for summarizing lecture slides to enhance students' preview efficiency and understanding of the content. Students are often asked to prepare for a class by reading lecture materials. However, because the attention span of students is limited, this is not always beneficial. We surveyed 326 students regarding the preview of lecture materials, revealing a preference for summarized materials to preview. Therefore, we developed an automatic summarization method for condensing original lecture materials into a summarized set. Our proposed approach utilizes image and text processing to extract important pages from lecture materials, optimizing selection of pages in accordance with a specified preview time. We applied the proposed summarization method to a set of lecture slides. In an experiment with 372 students, we compared the effectiveness of the summarized slides and the original materials in terms of quiz scores, preview achievement ratio, and time spent previewing. We found that students who previewed the summarized slides achieved better scores on pre-lecture quizzes, even though they spent less time previewing the material.

The present paper proposes and evaluates a readability assessment method designed for Japanese learners of EFL (English as a foreign language). The proposed readability assessment method is constructed by a regression algorithm using a new set of linguistic features that were employed separately in previous studies. The results showed that the proposed readability assessment method, which used all the linguistic features employed in previous studies, yielded a lower error of assessment than readability assessment methods using only some of these linguistic features. (Contains 4 figures and 2 tables.)

Increasing needs for real-time flood forecasting, proactive disaster prevention using hazard maps, and adaptation to flood risks associated with climate change require more detailed and accurate inundation information. Although river levees are an important factor in defining the extent and depth of flood water, data on the height and location of levees that can be introduced into global river models are not well developed. Therefore, in this study, an algorithm was developed to automatically determine the presence of levees when multiple river levee conditions are met from high-resolution Digital Elevation Models (DEMs). In the case study on the Kinu river, the 5-meter resolution DEM data was used to properly extract the location and height of levees including discontinuous levees, and the average error in levee height was in the range of 0.7 m. When a 10-meter, 20-meter, and 30-meter resolution DEM was used, the levee location was detected reasonably while erroneous determinations increased as the resolution became coarser, suggesting that the automatic detection method requires a resolution of at least 10-meter. [ABSTRACT FROM AUTHOR]

Labour shortages and fatal accidents in agricultural work have recently emerged as critical problems in Japan, necessitating productivity enhancement, workload reduction, and safety assurance. Therefore, in Japan and countries with similar agricultural environments, the use of small and inexpensive agricultural robots that can be employed in mountain farms and orchards is desirable. In this study, a dynamic positioning test was performed in orchards in a mountainous region to verify the positioning accuracy and stability of the global navigation satellite system (GNSS) and realtime kinematic (RTK)-GNSS. In addition, a simulator for an agricultural robot that could consider the environmental information of orchards was developed, and driving tests were conducted using the GNSS data acquired in the simulation. The error of the GNSS module was set to be higher than that for the measured value, and the robot travelling in the orchard was simulated. The results of GNSS positioning tests in an orchard near a mountainous area indicate that in the specific environmental conditions, the RTK-GNSS and stand-alone (SA)-GNSS can attain a positioning accuracy with an order of tens of centimetres and few metres, respectively. Moreover, the simulation results based on the GNSS positioning results indicate that a vehicle implementing RTKGNSS and a simple obstacle detection sensor can travel autonomously in a farmyard without colliding with the tree rows. In contrast, a vehicle implementing SA-GNSS and a simple obstacle detection sensor cannot drive autonomously in an orchard and must realise self-positioning using a more accurate sensor. Therefore, the proposed approach of realising simulations of autonomous agricultural robots based on GNSS data from a real orchard can facilitate the evaluation of practical agricultural robots and confirm safe travelling roots. Furthermore, the results demonstrate the possibility of developing small agricultural robots for orchards. We conducted the GNSS positioning test in an orchard at an altitude of approximately 830 m. A similar performance can be expected under alike agricultural situations because the error of the GNSS module was set to be higher than the measured value in the driving simulation test. [ABSTRACT FROM AUTHOR]

Extensive floating macroalgae have drifted from the East China Sea to Japan's offshore area, and field observation cannot sufficiently grasp their extensive spatial and temporal changes. High-spatial-resolution satellite data, which contain multiple spectral bands, have advanced remote sensing analysis. Several indexes for recognizing vegetation in satellite images, namely, the normalized difference vegetation index (NDVI), normalized difference water index (NDWI), and floating algae index (FAI), are useful for detecting floating macroalgae. Thresholds are defined to separate macroalgae-containing image pixels from other pixels, and adaptive thresholding increases the reliability of image segmentation. This study proposes adaptive thresholding using Sentinel-2 satellite data with a 10 m spatial resolution. We compare the abilities of Otsu's, exclusion, and standard deviation methods to define the floating macroalgae detection thresholds of NDVI, NDWI, and FAI images. This comparison determines the most advantageous method for the automatic detection of floating macroalgae. Finally, the spatial coverage of floating macroalgae and the reproducible combination needed for the automatic detection of floating macroalgae in Kagoshima, Japan, are examined. [ABSTRACT FROM AUTHOR]

SUMMARY In recent years, the number of vehicles with automatic gap control (AGC) (e.g., adaptive cruise control) has increased with the development of vehicle control technology. AGC mainly helps to ease driver fatigue on highways. Vehicles equipped with AGC can maintain an appropriate distance between two vehicles. Therefore, the traffic flow can be improved by AGC-equipped vehicles. In this research, we develop a traffic flow model that considers AGC-equipped vehicles, and is based on the Nagel-Schreckenberg model, which is a probabilistic cellular automaton model. The Nagel-Schreckenberg model can reproduce the basic features of traffic flow. On analyzing the mathematical model for the traffic including AGC-equipped vehicles, we can obtain analytical solutions of the traffic flow and the average velocity. [ABSTRACT FROM AUTHOR]

Time-saving technologies can raise people's standard of living by increasing their amount of disposable time. However, this may lead t the time-use rebound effect (TRE), wherein people reallocate their regained free time into actions that could affect energy consumption and greenhouse gas (GHG) emissions. In this empirical study, we used a causal inference framework to test whether TRE occurs by conducting a distribution experiment of time-saving goods. We obtained data from a questionnaire about the daily time use of the households in our experiment. Our results showed that the random assignment of time-saving goods leads to a time reallocation of household behaviour. These goods increased electricity use, although their magnitude is relatively small at 1.83% of household electricity consumption. • Time-saving technologies like automatic vacuum cleaners affect time and energy use. • Their impact on energy via behavioural change is the time-use rebound effect (TRE). • We investigate the TRE by randomly distributing automatic vacuum cleaners. • We found that automatic vacuum cleaners generated a TRE in households. • TRE is relatively small at 1.83% of total household electricity consumption. [ABSTRACT FROM AUTHOR]

SUMMARY The adoption of train automatic stop control (TASC) device is expected to increase according to high demand of platform door system. The challenges of TASC are (1) reduction of adjustment man-hour at TASC introduction and (2) maintenance of stopping position accuracy. These challenges are caused by the difficulty of obtaining vehicle characteristics accurately in real time. To solve the difficulty, a function for automatic parameter tuning for TASC devices has been developed. The function learns vehicle characteristics from real-time driving data and controls vehicle deceleration based on the learned vehicle characteristics. The developed function was evaluated on a test vehicle, and the results show that it acquired vehicle characteristics accurately and stopped the test vehicle within a stopping accuracy of +/- 35 cm. In addition, it was confirmed that the calculation time was within 70 s. This was less than the target time of 80 s, which is comparable to the average driving time between two stations on a metro line. [ABSTRACT FROM AUTHOR]

Simple Summary: We aimed to develop and evaluate an automatic prediction system for grading histopathological images of prostate cancer using a deep learning model and label distribution learning. Our results show that the label distribution learning improved the diagnostic performance of the automatic prediction system for the cancer grading. We aimed to develop and evaluate an automatic prediction system for grading histopathological images of prostate cancer. A total of 10,616 whole slide images (WSIs) of prostate tissue were used in this study. The WSIs from one institution (5160 WSIs) were used as the development set, while those from the other institution (5456 WSIs) were used as the unseen test set. Label distribution learning (LDL) was used to address a difference in label characteristics between the development and test sets. A combination of EfficientNet (a deep learning model) and LDL was utilized to develop an automatic prediction system. Quadratic weighted kappa (QWK) and accuracy in the test set were used as the evaluation metrics. The QWK and accuracy were compared between systems with and without LDL to evaluate the usefulness of LDL in system development. The QWK and accuracy were 0.364 and 0.407 in the systems with LDL and 0.240 and 0.247 in those without LDL, respectively. Thus, LDL improved the diagnostic performance of the automatic prediction system for the grading of histopathological images for cancer. By handling the difference in label characteristics using LDL, the diagnostic performance of the automatic prediction system could be improved for prostate cancer grading. [ABSTRACT FROM AUTHOR]

The aorta, including the thoracic and abdominal aortas, distributes blood from the heart to the upper and lower circulatory systems. Abdominal aortic sclerosis can cause abdominal aortic aneurysm. When the diameter of the aorta increases by more than 1.5 times its normal size (e.g., ≥4−5 cm, mean diameter in an adult is 2−3 cm), the risk of aneurysm rupture gradually increases. Aneurysm rupture and its associated bleeding have high mortality rates. Hence, the prediction of large diameter aortic aneurysms is an important issue with great clinical implications. In computed tomography angiography (CTA) imaging examination, an assistive tool for automatic image segmentation must be developed to help cardiologists and radiologists evaluate the risk of rupture progression and make appropriate treatment decisions. In this study, two‐dimensional (2D) spatial convolution operations (corner detection) and a multilayer machine vision classifier enabled the automatic segmentation of the luminal region, vascular intima, and calcified plaques in the abdominal aorta. The 2D spatial convolution operations can efficiently increase focus, remove unwanted noise in the convolutional layer, and segment calcified plaques and the vessel border by using a general regression neural network (GRNN)‐based classifier. The classifier's outputs can also be visualized in color presentation (red, green, and blue color series) to identify vessel borders and calcified plaques. Luminal change rate (LCR), calcified plaque rate (CPR), and ellipse ratio (ER) can be defined as criteria for evaluating changes in the luminal region and locating calcified plaques, which can be used to identify the locations of cross‐sectional shape changes and calcified plaques. Based on the collected CTA images from the clinical case studies, the proposed multilayer classifier has promising results for the intended medical purpose. With the LCR index, the correlation coefficients (R2) for manual image segmentation and automatic image segmentation is >0.90 in positive correlation for vessel borders segmentation. The proposed method is an adaptive learning scheme with potential clinical applications in automatic CTA image segmentation for further abdominal aortic aneurysm (AAA) risk evaluation. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Nowadays, fish farming plays an essential role in aquaculture. In the process of fry sales, the traditional manual counting method takes a great deal of time. Because the fish are constantly moving in the water, they are difficult to count and are likely to be unnecessarily damaged. With the rise of computer vision, fry counting based on image processing is also developing. It has the advantages of having high accuracy and fast processing speed. The primary counting process is divided into detection, adhesion segmentation, and fry counting. First, we improved the maximum interclass variance threshold selection method, and we extracted the histogram features under different channels to detect the fry; then, a matching segmentation method based on concave and skeleton crossing points was used to segment the adherent fry to improve the counting accuracy, the concaves were detected by CNN. We tested the method on real fry data sets, and the average accuracy reached 97%. Our approach had better accuracy than any existing current fish counting methods. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In this study, the viral genome extraction performance of automatic nucleic acid extractors and manual nucleic acid extraction kits was compared. We showed that compared with manual kits, the automatic extractors showed superior genome extraction performance using bovine viral diarrhea virus (BVDV) genome-positive cattle sera and bovine coronavirus/infectious bovine rhinotracheitis virus-spiked cattle nasal swabs. In addition, the subgenotyping of BVDV strains detected in Tokachi Province in Japan during 2016-2017 was performed. Results showed that most of these BVDV strains belonged to subgenotype 1b, while few strains belonged to subgenotypes 1a and 2a. This study showed the high applicability of automatic nucleic acid extractors in extracting multiple viral genomes and the dominant subgenotype of BVDV in Tokachi.

Background: The semi‐automatic intraosseous device (EZ‐IO®) is useful for safely securing intraosseous access. There are some complications associated with its use but few studies have examined the risk factors. The present study aimed to investigate the risk factors for extravasation as a complication of EZ‐IO use in pediatric patients. Methods: This study is a secondary analysis of a previous, monocentric, retrospective study conducted in Japan describing the use of EZ‐IO in a pediatric emergency department. The patients included in the study were younger than 16 years of age with EZ‐IO use at the Tokyo Metropolitan Children's Medical Center between January, 2013 and August, 2018. The factors analyzed included demographic information (sex, age, weight), Glasgow Coma Scale (GCS), diagnosis, bolus infusion, and lidocaine use. The primary endpoint was the odds ratio (OR) and 95% confidence interval (CI) for extravasation incidence. Results: Seventy‐two patients were enrolled; 14 of these had extravasation, 39 (54.2%) had a diagnosis of cardiac arrest, and 50 (69.4%) had a GCS score of 8 or less. Statistical analysis demonstrated that the group with a high GCS score was significantly associated with extravasation: GCS (<13) versus GCS (13≦) (OR: 12.25; 95% CI: 2.54–59.15; p < 0.01), GCS (<8) versus GCS (8≦) (OR: 4.78; 95% CI: 1.34–17.01; p = 0.03). Conclusions: A high GCS score was associated with extravasation in EZ‐IO use significantly more often than a low GCS score. No significant difference was found in the other endpoints. [ABSTRACT FROM AUTHOR]

Accurately‐labeled data is crucial for the training of machine learning models. For singing‐related tasks in the music information retrieval field, accurately‐labeled data is limited because annotating singing is time‐consuming. Several studies create vocal datasets using a two‐step annotation method which creates coarse labels first and then executes a manual calibration procedure. However, manually calibrating coarsely‐labeled singing data is expensive and time‐consuming. To address this problem, in this study we propose a singing‐label calibration framework, which aims to automatically calibrate the coarsely‐labeled singing data with higher accuracy. This framework contains a data augmentation method to generate training and testing data, a reasonable data preprocessing method to handle music audio and symbolic labels, a fully‐convolutional neural network to estimate the difference between coarse labels and accurate labels, and a novel calibration function to correct the coarse labels. Various experiments are conducted to examine the effect of our research. The results show that our model can highly reduce the cost time and slightly increase the labeling accuracy of the manual calibration process. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

This article describes a new automatic digital content generation system we have developed. Recently some universities, including Hosei University, have been offering students opportunities to take distance interactive classes over the Internet from overseas. When such distance lectures are delivered in English to Japanese students, there is a pressing need to provide materials for review after class, such as video content on a CD-ROM or on a Web site. To meet this need, we have developed a new automatic content generation system, which enables the complete archiving of lectures including video/audio content, synchronized presentation materials, and handwritten traces on virtual whiteboards. The content is generated in real time and is immediately available at the end of the class. In addition, this system incorporates a unique video search algorithm which adopts a phonetic-based search technology. This enables quick review of the video content by typed-in keywords. The system can automatically create a vast amount of digital content and provide students with an efficient learning tool.

The uncertainty in walking and waiting transfer times at transit stops depends on the environmental conditions, vehicle schedule punctuality, and variations among individuals. Relying only on the mean values ignores such uncertainty and leads to improper transfer coordination. This paper provides a general framework to estimate the time distribution parameters of walking and waiting based on automatic-fare-collection data under the assumption that the distribution function is known. The proposed methodology contributes to the existing literature by providing a stochastic frontier model (SFM) that is adaptable to transit assignment models. The normal/exponential SFM is utilized to estimate distribution parameters based on the maximum-likelihood method. The methodology was applied to smart-card data from Japan to estimate the transfer walking- and waiting-time distributions between bus stops and rail stations. The results indicate a low disturbance in the observed data for walking time and that our assumptions regarding walking-time estimation were reliable. Furthermore, we quantified how the waiting time at a bus stop depends on variations in the service headway. These findings are expected to help in making more informed decisions to optimize transit schedules. [ABSTRACT FROM AUTHOR]

Most studies on the Stroop effect (unintentional automatic word processing) have been restricted to English speakers using vocal responses. Little is known about this effect with deaf signers. The study compared Stroop task responses among four different samples: deaf participants from a Japanese-language environment and from an English-language environment; and hearing individuals from Japan and from Australia. Color words were prepared in both English and Japanese and were presented in three conditions: congruent (e.g., the word "red" printed in red), incongruent (e.g., "red" printed in blue), and neutral. The magnitude of the effect was greater with the deaf participants than with the hearing participants. The deaf individuals experienced more interference in English than in Japanese. (Contains 2 tables.)

Automatic individual recognition techniques can support data collection in the field of ethology. Recent studies have contributed to development of automatic individual recognition techniques using machine learning and deep learning. However, varied conditions in the wild, such as the presence of occlusions and head rotations of individuals, can lower the accuracy of automatic recognition techniques. Thus, there is requirement for improvement in the accuracy and robustness of these techniques. In this study, we have used previously observed information updated with given current observation by Bayesian inference to improve the automatic individual recognition of free‐ranging Japanese macaques (Macaca fuscata) at Katsuyama, Japan. We collected static images and video footage of 51 adult individuals. Using the static images, we created eight individual recognition systems (classifiers), using GoogLeNet and ResNet‐18 as convolutional neural network models. Additionally, sequential data of the faces of the macaques were obtained from 86 video recordings of the 51 individuals to evaluate the classifiers. We were able to successfully recognize 90% or more individuals with each classifier through the combination of the sequential Bayesian filter and the classifier. Eighty‐five percent or more of the individuals had posterior probabilities of 90% or above when conducting recognition tests using the sequential Bayesian filter with 10 images. The best classifier recognized 98% of individuals using 10 images and all individuals using 50 images. Recognition was also successful when the sequential Bayesian filter was applied to cases in which the recognition rate was <50% with test data when the filter was not applied. Based on the above results, we posit that the accuracy of individual recognition systems can be improved using a sequential Bayesian filter that considers past information for individuals. [ABSTRACT FROM AUTHOR]

Background/Objectives: Artificial Intelligence (AI) technology is gaining attention for its potential in cognitive function assessment and intervention. AI robots and agents can offer continuous dialogue with the elderly, helping to prevent social isolation and support cognitive health. Speech-based evaluation methods are promising as they reduce the burden on elderly participants. AI agents could replace human questioners, offering efficient and consistent assessments. However, existing research lacks sufficient comparisons of elderly speech content when interacting with AI versus human partners, and detailed analyses of factors like cognitive function levels and dialogue partner effects on speech elements such as proper nouns and fillers. Methods: This study investigates how elderly individuals' cognitive functions influence their communication patterns with both human and AI conversational partners. A total of 34 older people (12 men and 22 women) living in the community were selected from a silver human resource centre and day service centre in Tokyo. Cognitive function was assessed using the Mini-Mental State Examination (MMSE), and participants engaged in semi-structured daily conversations with both human and AI partners. Results: The study examined the frequency of fillers, proper nouns, and "listen back" in conversations with AI and humans. Results showed that participants used more fillers in human conversations, especially those with lower cognitive function. In contrast, proper nouns were used more in AI conversations, particularly by those with higher cognitive function. Participants also asked for explanations more often in AI conversations, especially those with lower cognitive function. These findings highlight differences in conversation patterns based on cognitive function and the conversation partner being either AI or human. Conclusions: These results suggest that there are differences in conversation patterns depending on the cognitive function of the participants and whether the conversation partner is a human or an AI. This study aims to provide new insights into the effective use of AI agents in dialogue with the elderly, contributing to the improvement of elderly welfare. [ABSTRACT FROM AUTHOR]

In this study, we focus on ethical education as a means to improve artificial companion's conceptualization of moral decision-making process in human users. In particular, we focus on automatically determining whether changes in ethical education influenced core moral values in humans throughout the century. We analyze ethics as taught in Japan before WWII and today to verify how much the pre-WWII moral attitudes have in common with those of contemporary Japanese, to what degree what is taught as ethics in school overlaps with the general population's understanding of ethics, as well as to verify whether a major reform of the guidelines for teaching the school subject of "ethics" at school after 1946 has changed the way common people approach core moral questions (such as those concerning the sacredness of human life). We selected textbooks used in teaching ethics at school from between 1935 and 1937, and those used in junior high schools today (2019) and analyzed what emotional and moral associations such contents generated. The analysis was performed with an automatic moral and emotional reasoning agent and based on the largest available text corpus in Japanese as well as on the resources of a Japanese digital library. As a result, we found out that, despite changes in stereotypical view on Japan's moral sentiments, especially due to historical events, past and contemporary Japanese share a similar moral evaluation of certain basic moral concepts, although there is a large discrepancy between how they perceive some actions to be beneficial to the society as a whole while at the same time being inconclusive when it comes to assessing the same action's outcome on the individual performing them and in terms of emotional consequences. Some ethical categories, assessed positively before the war, while being associated with a nationalistic trend in education have also disappeared from the scope of interest of post- war society. The findings of this study support suggestions proposed by others that the development of personal AI systems requires supplementation with moral reasoning. Moreover, the paper builds upon this idea and further suggests that AI systems need to be aware of ethics not as a constant, but as a function with a correction on historical and cultural changes in moral reasoning. [ABSTRACT FROM AUTHOR]

Objective: To evaluate the capacity of the automatic detection system to accurately grade, from smartphones' selfie pictures, the severity of ten facial signs in Japanese women and their changes due to age and sun exposures., Methods: A three-step approach was conducted, based on self-taken selfie images. At first, to check on 310 Japanese women (18-69 years) enrolled in the northerner Hokkaido area (latitude 43.2°N), how, on ten facial signs, the A.I-based automatic grading system may correlate with dermatological assessments, taken as reference. Second, to assess and compare age changes in 310 Japanese and 112 Korean women. Third, as these Japanese panelists were recruited according to their usual behavior toward sun exposure, that is, non-sun-phobic (NSP, N = 114) and sun-phobic (SP, N = 196), and through their regular and early use of a photo-protective product, to characterize the facial photo-damages., Results: (a) On the ten facial signs, detected automatically, nine were found significantly (P < .0001) highly correlated with the evaluations made by three Japanese dermatologists (Wrinkles: r = .75; Sagging: r = .80; Pigmentation: r = .75). (b) The automatic scores showed significant changes with age, by decade, of Wrinkles/Texture, Pigmentation, and Ptosis/Sagging (P < .05). (c) After 45 years, a significantly increased severity of Wrinkles/Texture and Pigmentation was observed in NSP vs. SP women (P < .05). A trend of an increased Ptosis/Sagging (P = .09) was observed., Conclusion: This work illustrates, for the first time through investigations conducted at home, some impacts of aging and sun exposures on facial signs of Japanese women. Results significantly confirm the importance of sun avoidance coupled with photo-protective measures. In epidemiological studies, the AI-based system offers a fast, affordable, and confidential approach in detection and quantification of facial signs and their dependence with ages, environments and lifestyles., (© 2020 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.)

Introduction: To increase survival rates of patients with severe trauma from road traffic accidents, Japan launched the D-Call Net (DCN) system, which utilizes advanced automatic collision notification technology to dispatch doctors by helicopter. The DCN system began in November 2015 and, as of October 2019, has dispatched doctors 4 times. Cases: Case 1-Canceled because trauma was mild. Case 2-Doctor made contact with 2 patients with moderate trauma 29 minutes earlier than would have occurred conventionally. This was the first case in the world to use automotive engineering data to dispatch a doctor to a patient. Case 3-An accident involving 3 severely injured patients activated DCN, enabling doctor-patient contact 20 minutes earlier than would have been possible conventionally. Case 4-DCN was ineffective. Discussion: According to 2008 data from Chiba Prefecture, in accidents where victims sustain severe trauma, the interval from accident occurrence to hospital arrival was 67 minutes, even when doctors were dispatched by air ambulance (Doctor-Heli [DH]). Use of accident information for faster doctor dispatch effectively improved survival rates. An algorithm was developed to use accident information to assess trauma severity (severity probability). DCN dispatches doctors by using data, including information on accident site and severity probability, which are sent to smartphones of doctors, thereby reducing the interval from accident to DH request by approximately 17 minutes. DCN is the first system in the world to use automotive engineering information for faster doctor dispatch to traffic accident sites. The system is crucial for improving survival rates and mitigating the aftereffects of traffic accidents. [ABSTRACT FROM AUTHOR]

In Japan, nursing records are not easily put to secondary use because nursing documentation is not standardized. In recent years, electronic health records have necessitated the creation of Japanese nursing terminology. The purpose of this study was to develop and evaluate an automatic classification system for narrative nursing records using natural language processing technology and machine learning. We collected a week's worth of narrative nursing records from an academic hospital. The authors independently annotated the text data, dividing it into morphemes, the smallest meaningful unit in a language. During preprocessing when creating feature quantities, we used a Japanese tokenizer, MeCab, an open-source morphological parser, and the bag-of-words model. A support vector machine was adopted as a classifier for machine learning. The accuracy was 0.96 and 0.86 on the training set and test set, respectively, and the F value was 0.82. Our findings provide useful information regarding the development of an automatic classification system for Japanese nursing records using nursing terminology and natural language processing techniques., (Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.)

Detecting and tracking objects in low Earth orbit is an increasingly important task. Telescope observations contribute to its accomplishment, and telescope imagers produce a large amount of data for this task. Thus, it is convenient to use fast computer-aided processes to analyze it. Telescopes tracking at the sidereal rate usually detect these objects in their imagers as streaks, their lengths depending on the exposure time and the slant range to the object. We have developed a processing pipeline to automatically detect streaks in astronomical images in real time (i.e., faster than the images are produced) by a graphics processing unit parallel processing system. After the detection stage, streak photometric information is obtained, and object candidate identification is provided through matches with a two-line element set database. The system has been tested on a large set of images, consisting of two hours of observation time, from the Tomo-e Gozen camera of the 105 cm Schmidt telescope at Kiso Observatory in Japan. Streaks were automatically detected in approximately 0.5% of the images. The process detected streaks down to a minimum apparent magnitude of +11.3 and matched the streaks with objects from the space-track catalog in 78% of the cases. We believe that this processing pipeline can be instrumental in detecting new objects and tracking existing ones when processing speed is important, for instance, when a short handover time is required between follow-up observation stations, or when there is a large number of images to process. This study will contribute to consolidating optical observations as an effective way to control and alleviate the space debris problem. [ABSTRACT FROM AUTHOR]

The use of geared motor modules, installed in automatic doors to optimize the stability, efficiency, and manufacturing costs related to automatic door opening and closing, has been attracting scientific attention. Generally, signal processing and the sensor circuit board inserted in the geared motor have been widely used in optical systems. However, the optical system is disadvantageous in that it is susceptible to external shock, humidity, and dust. In this article, a magnetic impulse ring‐based motion detection circuit board simpler in structure than the existing optical systems, having excellent durability against external environments, and can be miniaturized is proposed. Experimental results show that the proposed module based on Hall sensor has RPM (revolution per minute) detection characteristics more than 6 dB better than conventional optical module. Also, the proposed module has about 14 dB excellent power consumption than existing optical modules, thus its applicability is confirmed. © 2019 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. [ABSTRACT FROM AUTHOR]

Abstract: PON (Passive Optical Network) permits the economical deployment of FTTH (Fiber to the Home) networks by sharing a single optical subscriber line among multiple subscribers. Especially since 2002, the deployment of PON systems has grown rapidly. This paper introduces the current status of FTTH in Japan then proposes a redundant optical access system with APS (Automatic Protection Switching) over PON which enables PONs for FTTBus (Fiber to the Business). [Copyright &y& Elsevier]

Underwater image segmentation is useful for benthic habitat mapping and monitoring; however, manual annotation is time-consuming and tedious. We propose automated segmentation of benthic habitats using unsupervised semantic algorithms. Four such algorithms––Fast and Robust Fuzzy C-Means (FR), Superpixel-Based Fast Fuzzy C-Means (FF), Otsu clustering (OS), and K-means segmentation (KM)––were tested for accuracy for segmentation. Further, YCbCr and the Commission Internationale de l'Éclairage (CIE) LAB color spaces were evaluated to correct variations in image illumination and shadow effects. Benthic habitat field data from a geo-located high-resolution towed camera were used to evaluate proposed algorithms. The Shiraho study area, located off Ishigaki Island, Japan, was used, and six benthic habitats were classified. These categories were corals (Acropora and Porites), blue corals (Heliopora coerulea), brown algae, other algae, sediments, and seagrass (Thalassia hemprichii). Analysis showed that the K-means clustering algorithm yielded the highest overall accuracy. However, the differences between the KM and OS overall accuracies were statistically insignificant at the 5% level. Findings showed the importance of eliminating underwater illumination variations and outperformance of the red difference chrominance values (Cr) in the YCbCr color space for habitat segmentation. The proposed framework enhanced the automation of benthic habitat classification processes. [ABSTRACT FROM AUTHOR]

Importance: A high level of surgical skill is essential to prevent intraoperative problems. One important aspect of surgical education is surgical skill assessment, with pertinent feedback facilitating efficient skill acquisition by novices., Objectives: To develop a 3-dimensional (3-D) convolutional neural network (CNN) model for automatic surgical skill assessment and to evaluate the performance of the model in classification tasks by using laparoscopic colorectal surgical videos., Design, Setting, and Participants: This prognostic study used surgical videos acquired prior to 2017. In total, 650 laparoscopic colorectal surgical videos were provided for study purposes by the Japan Society for Endoscopic Surgery, and 74 were randomly extracted. Every video had highly reliable scores based on the Endoscopic Surgical Skill Qualification System (ESSQS, range 1-100, with higher scores indicating greater surgical skill) established by the society. Data were analyzed June to December 2020., Main Outcomes and Measures: From the groups with scores less than the difference between the mean and 2 SDs, within the range spanning the mean and 1 SD, and greater than the sum of the mean and 2 SDs, 17, 26, and 31 videos, respectively, were randomly extracted. In total, 1480 video clips with a length of 40 seconds each were extracted for each surgical step (medial mobilization, lateral mobilization, inferior mesenteric artery transection, and mesorectal transection) and separated into 1184 training sets and 296 test sets. Automatic surgical skill classification was performed based on spatiotemporal video analysis using the fully automated 3-D CNN model, and classification accuracies and screening accuracies for the groups with scores less than the mean minus 2 SDs and greater than the mean plus 2 SDs were calculated., Results: The mean (SD) ESSQS score of all 650 intraoperative videos was 66.2 (8.6) points and for the 74 videos used in the study, 67.6 (16.1) points. The proposed 3-D CNN model automatically classified video clips into groups with scores less than the mean minus 2 SDs, within 1 SD of the mean, and greater than the mean plus 2 SDs with a mean (SD) accuracy of 75.0% (6.3%). The highest accuracy was 83.8% for the inferior mesenteric artery transection. The model also screened for the group with scores less than the mean minus 2 SDs with 94.1% sensitivity and 96.5% specificity and for group with greater than the mean plus 2 SDs with 87.1% sensitivity and 86.0% specificity., Conclusions and Relevance: The results of this prognostic study showed that the proposed 3-D CNN model classified laparoscopic colorectal surgical videos with sufficient accuracy to be used for screening groups with scores greater than the mean plus 2 SDs and less than the mean minus 2 SDs. The proposed approach was fully automatic and easy to use for various types of surgery, and no special annotations or kinetics data extraction were required, indicating that this approach warrants further development for application to automatic surgical skill assessment.

The number of medical accidents has recently been increasing in patient environments in Japan. Most of these accidents are ascribed to errors by nurses. Among these, however, are cases that could have been prevented by appropriate management of medical articles. In the patient environment, the patient is placed at the center, and medical articles are in a dependency relation to the patient. An accident may arise if such a dependency is not fulfilled. This paper proposes the use of low-frequency-band RFID tags in order to monitor the patient environment. In the proposed method, tags are attached to the patient and all related medical articles. By monitoring the tags at any time by wireless communications, mistaken inclusion of articles related to other patients can be avoided. In this system, tags are automatically identified, which avoids placing an extra burden on the nurses and other medical professionals. In this study, the identification performance of RFID tags is improved and the feasibility of their use as a practical system is verified by an experiment in an environment simulating an actual site. © 2004 Wiley Periodicals, Inc. Syst Comp Jpn, 35(10): 74–82, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.10581 [ABSTRACT FROM AUTHOR]

Background: Change and sustain talks (negative and positive comments) on gambling have been relevant for determining gamblers' outcomes but they have not been used to clarify the abstinence process in anonymous gambler meetings., Objective: The aim of this study was to develop a change talk model for abstinence based on data extracted from web-based anonymous gambler chat meetings by using an automatic change talk classifier., Methods: This study used registry data from the internet. The author accessed web-based anonymous gambler chat meetings in Japan and sampled 1.63 million utterances (two-sentence texts) from 267 abstinent gamblers who have remained abstinent for at least three years and 1625 nonabstinent gamblers. The change talk classifier in this study automatically classified gamblers' utterances into change and sustain talks., Results: Abstinent gamblers showed higher proportions of change talks and lower probability of sustain talks compared with nonabstinent gamblers. The change talk model for abstinence, involving change and sustain talks, classified abstinent and nonabstinent gamblers through the use of a support vector machine with a radial basis kernel function. The model also indicated individual evaluation scores for abstinence and the ideal proportion of change talks for all participants according to their previous utterances., Conclusions: Abstinence likelihood among gamblers can be increased by providing personalized evaluation values and indicating the optimal proportion of change talks. Moreover, this may help to prevent severe mental, social, and financial problems caused by the gambling disorder., (©Kenji Yokotani. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 21.06.2021.)

This study investigates the medium-scale traveling ionospheric disturbances (MSTIDs) statistically at the low-latitude equatorial ionization anomaly (EIA) region in the northern hemisphere. We apply the automatic detection algorithm including the three-dimensional fast Fourier transform (3-D FFT) and support vector machine (SVM) on total electron content (TEC) observations, derived from a network of ground-based global navigation satellite system (GNSS) receivers in Taiwan (14.5° N geomagnetic latitude; 32.5° inclination), to identify MSTID from other waves or irregularity features. The obtained results are analyzed statistically to examine the behavior of low-latitude MSTIDs. Statistical results indicate the following characteristics. First, the southward (equatorward) MSTIDs are observed almost every day during 0800–2100 LT in Spring and Winter. At midnight, southward MSTIDs are more discernible in Summer and majority of them are propagating from Japan to Taiwan. Second, northward (poleward) MSTIDs are more frequently detected during 1200–2100 LT in Spring and Summer with the secondary peak of occurrence between day of year (DOY) 100–140 during 0000–0300 LT. The characteristics of the MSTIDs are interpreted with additional observations from radio occultation (RO) soundings of FORMOSAT-3/COSMIC as well as modeled atmospheric waves from the high-resolution Whole Atmosphere Community Climate Model (WACCM) suggesting that the nighttime MSTIDs in Summer is likely connected to the atmospheric gravity waves (AGWs). [ABSTRACT FROM AUTHOR]

Introduction: To increase survival rates of patients with severe trauma from road traffic accidents, Japan launched the D-Call Net (DCN) system, which utilizes advanced automatic collision notification technology to dispatch doctors by helicopter. The DCN system began in November 2015 and, as of October 2019, has dispatched doctors 4 times., Cases: Case 1-Canceled because trauma was mild. Case 2-Doctor made contact with 2 patients with moderate trauma 29 minutes earlier than would have occurred conventionally. This was the first case in the world to use automotive engineering data to dispatch a doctor to a patient. Case 3-An accident involving 3 severely injured patients activated DCN, enabling doctor-patient contact 20 minutes earlier than would have been possible conventionally. Case 4-DCN was ineffective., Discussion: According to 2008 data from Chiba Prefecture, in accidents where victims sustain severe trauma, the interval from accident occurrence to hospital arrival was 67 minutes, even when doctors were dispatched by air ambulance (Doctor-Heli [DH]). Use of accident information for faster doctor dispatch effectively improved survival rates. An algorithm was developed to use accident information to assess trauma severity (severity probability). DCN dispatches doctors by using data, including information on accident site and severity probability, which are sent to smartphones of doctors, thereby reducing the interval from accident to DH request by approximately 17 minutes. DCN is the first system in the world to use automotive engineering information for faster doctor dispatch to traffic accident sites. The system is crucial for improving survival rates and mitigating the aftereffects of traffic accidents.

Background: Keratometry is clinically important and is routinely performed as part of human ophthalmic examination. In veterinary ophthalmology, little is known about keratometry in dogs, and its practical application has been limited. The present study aimed to describe keratometry in some dog breeds popular in Japan using a handheld keratometer., Methods: Client-owned dogs of various signalment were enrolled prospectively in the keratometry examination. Interbreed variations in mean corneal curvatures (R1R2avg) and corneal astigmatism (Δ(R1-R2)) were evaluated statistically with respect to their bodyweight based on the data which fulfilled the predetermined inclusion criteria. P<0.05 was considered statistically significant., Results: On examination of 237 dogs from 16 different breeds, R1R2avg (mean±sd) ranged from 7.54±0.30 mm in Pomeranians to 9.28±0.19 mm in golden retrievers. Δ(R1-R2) (mean±sd) ranged from 0.22±0.11 mm in miniature schnauzers to 0.57±0.30 mm in French bulldogs., Conclusion: The present study successfully described keratometry in 16 dog breeds. The study revealed considerable interbreed variations in both R1R2avg and Δ(R1-R2), which did not necessarily correlate with bodyweight. These results are useful both clinically in fitting contact lenses in the management of corneal diseases and non-clinically in optometric studies in dogs., Competing Interests: Competing interests: None declared., (© British Veterinary Association 2020. Re-use permitted under CC BY-NC. No commercial re-use. Published by BMJ.)

• Automatic assessment of L2 phonetic fluency in read-aloud speech without an ASR system. • High agreement between automatic and human assessments. • Longitudinal coherence between fluency scores and learning stage. Automatic second language (L2) speech fluency assessment has been one of the ultimate goals of several projects aiming at designing Computer-Assisted Pronunciation Training (CAPT) tools for L2 learners. Usually, three challenges must be tackled in order to solve the issues at stake: 1) Defining fluency from a threefold interdisciplinary perspective (acoustic and perceptual phonetics, computer science, L2 education); 2) Using a cost-effective algorithm; 3) Testing the procedure with actual learners' data. Despite rapid technical developments in the field of automatic speech processing, the tools which are actually available for learners are still scarce, and most of them rely on automatic speech recognition (ASR). Moreover, most research on the topic is focusing on English as the target L2. Therefore, in this article, we address the following research questions: (a) is it possible to use a non-ASR-based low-level signal segmentation algorithm to predict human expert assessment of phonetic fluency in beginner Japanese learners of French in a text-reading task during the first stages of their learning? (b) if the answer to (a) is positive, then what are the best predictors of phonetic fluency among a set of available measures (see below for more details)? (c) is it possible to use this algorithm to monitor the evolution of phonetic fluency (and of its associated predictors) in these learners in a longitudinal study? As a first step, a corpus of French sentences read aloud by 12 Japanese learners of different proficiency levels in French was used to design a prediction system. The read-aloud speech data was perceptually annotated by three human experts on four dimensions: overall speech fluency, speech rate, regularity of speech rate, speech fluidity (i.e. smoothness of transitions between phones). Inter-rater agreement and reliability were high for all dimensions, and the average human ratings were compared with the scores provided by our prediction system. The results show strong correlations between human and automatic scores of speech rate and regularity of speech rate, and a weak correlation for speech fluidity. Automatic scores were finally combined together through a multiple linear regression model in order to predict overall speech fluency. The best model led to a correlation coefficient of.92 between automatic and human ratings, with a root-mean-square error of.38. In the second step of this study, a corpus of identical sentences read aloud four times over two years by 12 Japanese learners of French (after 4, 7, 12, and 19 months of French courses in Japan) was fed to the automatic system. The results show regular progress in overall speech fluency, which fits with the regular progress the Japanese learners under scrutiny were expected to make through their academic program in French at their university in Japan every semester. Our study suggests a positive answer to our first and third research questions, with speech rate as the best predictor to answer our second research question. In a pedagogical perspective, it seems that such a simple algorithm could be integrated in a CAPT tool to monitor learners' progress in phonetic fluency in reading-aloud tasks. [ABSTRACT FROM AUTHOR]

Fine-scale thermohaline structure within ocean column can be mapped seismically in the Kuroshio Current, off the Muroto Peninsula of Shikoku Island, Japan. In this paper, we present the application of automatic sound speed picking analysis to the multi-channel seismic reflection data acquired in a different period to estimate time-lapse sound speed distribution across the Kuroshio Current. This method is based on an optimal velocity trajectory solving by the eikonal equation with a finite-difference algorithm. In contrast to the seismic inversion technique, this automatic analysis enables us to obtain contrast sound speed profiles without heavy dependency on sound speed or temperature data directly measured at discrete locations. As a result, this method can visualise sound speed profiles of fine-scale thermohaline structure developed at interleaving or diapycnal mixing processes of different water masses in the Kuroshio Current. The images of all profiles mapped from automatic sound speed analysis distinguish water masses and their fine-scale internal structure such as cold and warm water eddies, thermohaline staircases and internal waves revealing acoustic contrasts at interfaces across where sound speed and temperature change. Applying our approach for individual seismic line acquired in different time-steps for 3D seismic data can provide time–space variant images of fine-scale thermohaline structure for studies of oceanographic processes as well as large-scale ocean current and climate systems. [ABSTRACT FROM AUTHOR]

Automatic and rapid imaging of earthquake rupture process is a difficult task but critical for postearthquake response and possible tsunami early warning. We have developed an automatic system for finite‐fault inversion based on the iterative deconvolution and stacking (IDS) approach, in which we screen the waveforms and update the fault size automatically. Strong‐motion seismograms are used in offline tests to invert for the rupture processes of 34 significant earthquakes in 2008–2017 in and around Japan with magnitudes MW ≥ 6.0 occurred onshore and MW ≥ 6.5 offshore. Rupture model of an earthquake, including moment magnitude, source time function, slip distribution, rupture velocity, and stress drop, is obtained in ~1.5 min on a desktop PC. The moment magnitudes we obtained are consistent with those published by other agencies. The slip distributions, rupture directions, and velocities also agree well with those released by the Japan Meteorological Agency. In all test cases, our results show that the automatic inversion of strong‐motion data can provide reliable major source parameters without manual revision, which demonstrates great potential for emergency response and tsunami early warning purposes. The automatic inversion system developed in this work can also be used to build catalogs of finite‐fault models for moderate to strong earthquakes in other areas where the moment tensor solutions and near‐field waveform data are available. Key Points: We have developed an automatic system for finite‐fault inversion based on the iterative deconvolution and stacking (IDS) techniqueOffline tests demonstrate that the rupture models can be well estimated when the strong‐motion data and the mechanism solution are availableIntegration of the automatic finite‐fault inversion in a rapid response and early warning system is realistic for regions like Japan [ABSTRACT FROM AUTHOR]

In recent years, smart learning environments have become central to modern education and support students and instructors through tools based on prediction and recommendation models. These methods often use learning material metadata, such as the knowledge contained in an exercise which is usually labeled by domain experts and is costly and difficult to scale. It recognizes that automated labeling eases the workload on experts, as seen in previous studies using automatic classification algorithms for research papers and Japanese mathematical exercises. However, these studies didn't delve into fine-grained labeling. In addition to that, as the use of materials in the system becomes more widespread, paper materials are transformed into PDF formats, which can lead to incomplete extraction. However, there is less emphasis on labeling incomplete mathematical sentences to tackle this problem in the previous research. This study aims to achieve precise automated classification even from incomplete text inputs. To tackle these challenges, we propose a mathematical exercise labeling algorithm that can handle detailed labels, even for incomplete sentences, using word n-grams, compared to the state-of-the-art word embedding method. The results of the experiment show that mono-gram features with Random Forest models achieved the best performance with a macro F-measure of 92.50%, 61.28% for 24-class labeling and 297-class labeling tasks, respectively. The contribution of this research is showing that the proposed method based on traditional simple n-grams has the ability to find context-independent similarities in incomplete sentences and outperforms state-of-the-art word embedding methods in specific tasks like classifying short and incomplete texts. [ABSTRACT FROM AUTHOR]