Search

In recent years, automated analyses using novel NLP methods have been used to investigate language abnormalities in schizophrenia. In contrast, only a few studies used automated language analyses in bipolar disorder. To our knowledge, no previous research compared automated language characteristics of first-episode psychosis (FEP) and bipolar disorder (FEBD) using NLP methods. Our study included 53 FEP, 40 FEBD and 50 healthy control participants who are native Turkish speakers. Speech samples of the participants in the Thematic Apperception Test (TAT) underwent automated generic and part-of-speech analyses, as well as sentence-level semantic similarity analysis based on SBERT. Both FEBD and FEP were associated with the use of shorter sentences and increased sentence-level semantic similarity but less semantic alignment with the TAT pictures. FEP also demonstrated reduced verbosity and syntactic complexity. FEP differed from FEBD in reduced verbosity, decreased first-person singular pronouns, fewer conjunctions, increased semantic similarity as well as shorter sentence and word length. The mean classification accuracy was 82.45 % in FEP vs HC, 71.1 % in FEBD vs HC, and 73 % in FEP vs FEBD. After Bonferroni correction, the severity of negative symptoms in FEP was associated with reduced verbal output and increased 5th percentile of semantic similarity. The main limitation of this study was the cross-sectional nature. Our findings demonstrate that both patient groups showed language abnormalities, which were more severe and widespread in FEP compared to FEBD. Our results suggest that NLP methods reveal transdiagnostic linguistic abnormalities in FEP and FEBD. • Both FEP and FEBD demonstrated increased sentence-level semantic similarity. • FEP used shorter sentences and words compared to FEBD. • FEP also showed increased semantic similarity compared to FEBD. • The mean accuracy of the machine learning classifier was 73 % for FEP vs FEBD. • FEP exhibited more extensive and severe language abnormalities compared to FEBD. [ABSTRACT FROM AUTHOR]

Automatic spelling correction stands as a pivotal challenge within the ambit of natural language processing (NLP), demanding nuanced solutions. Traditional spelling correction techniques are typically only capable of detecting and correcting non-word errors, such as typos and misspellings. However, context-sensitive errors, also known as real-word errors, are more challenging to detect because they are valid words that are used incorrectly in a given context. The Persian language, characterized by its rich morphology and complex syntax, presents formidable challenges to automatic spelling correction systems. Furthermore, the limited availability of Persian language resources makes it difficult to train effective spelling correction models. This paper introduces a cutting-edge approach for precise and efficient real-word error correction in Persian text. Our methodology adopts a structured, multi-tiered approach, employing semantic analysis, feature selection, and advanced classifiers to enhance error detection and correction efficacy. The innovative architecture discovers and stores semantic similarities between words and phrases in Persian text. The classifiers accurately identify real-word errors, while the semantic ranking algorithm determines the most probable corrections for real-word errors, taking into account specific spelling correction and context properties such as context, semantic similarity, and edit-distance measures. Evaluations have demonstrated that our proposed method surpasses previous Persian real-word error correction models. Our method achieves an impressive F-measure of 96.6% in the detection phase and an accuracy of 99.1% in the correction phase. These results clearly indicate that our approach is a highly promising solution for automatic real-word error correction in Persian text. [ABSTRACT FROM AUTHOR]

With the growing number of genomic data in public repositories, efficient search methodologies have become a basic need to reach the relevant genomic data. However, this need cannot be fulfilled with the current repositories because they offer a limited search option which is a lexical matching of textual descriptions or metadata of the experiments. This technique is insufficient to get the required information needed to detect similarities between experiments within a large data collection. Due to the limitation of the existing repositories, in this study, we develop a text-based experiment retrieval framework by using both lexical and semantic similarity approaches to find similarities between experiments, and their retrieval performance was compared. This study is the first attempt to use text-driven semantic analysis approaches for developing a retrieval framework for experiments. An empirical study was conducted on a large textual description of Arabidopsis microarray experiments from the Gene Expression Omnibus database. In the proposed model, Jaccard similarity was used as a lexical similarity approach; Latent Semantic Analysis, Probabilistic Latent Semantic Analysis and Latent Dirichlet allocation were used as semantic similarity approaches to detect similarities between the textual descriptions of the experiments. According to the experimental results, relevant experiments can be retrieved successfully by text-driven semantic similarity approaches compared with the lexical similarity approach. [ABSTRACT FROM AUTHOR]

The co-opinionatedness measure, that is, the similarity of cociting documents in their opinions about their cocited articles, has been recently proposed. The present study uses a wider range of baselines and benchmarks to investigate the measure's effectiveness in retrieval ranking that was previously confirmed in a pilot study. A test collection was built including 30 seed documents and their 4702 cocited articles. Their citances and full-texts were analysed using natural language processing (NLP) and opinion mining techniques. Cocitation values, syntactical similarity and contexts similarity were used as baselines. The distributional semantic similarity and the linear and hierarchical Medical Subject Headings (MeSH) similarities served as benchmarks to evaluate the effect of the co-opinionatedness as a boosting factor on the performance of the baselines. The improvements in the rankings were measured by normalised discounted cumulative gain (nDCG). According to the findings, there existed significant differences between the nDCG mean values obtained before and after weighting the baselines by the co-opinionatedness measures. The results of the generalisability study corroborated the reliability and generalisability of the systems. Accordingly, the similarity in the opinions of the cociting papers towards their cocited articles can explain the cocitation relation in the scientific papers network and can be effectively utilised for improving the results of the cocitation-based retrieval systems. [ABSTRACT FROM AUTHOR]

Pre-trained transformer models shine in many natural language processing tasks and therefore are expected to bear the representation of the input sentence or text meaning. These sentence-level embeddings are also important in retrieval-augmented generation. But do commonly used plain averaging or prompt templates sufficiently capture and represent the underlying meaning? After providing a comprehensive review of existing sentence embedding extraction and refinement methods, we thoroughly test different combinations and our original extensions of the most promising ones on pretrained models. Namely, given 110 M parameters, BERT's hidden representations from multiple layers, and many tokens, we try diverse ways to extract optimal sentence embeddings. We test various token aggregation and representation post-processing techniques. We also test multiple ways of using a general Wikitext dataset to complement BERT's sentence embeddings. All methods are tested on eight Semantic Textual Similarity (STS), six short text clustering, and twelve classification tasks. We also evaluate our representation-shaping techniques on other static models, including random token representations. Proposed representation extraction methods improve the performance on STS and clustering tasks for all models considered. Very high improvements for static token-based models, especially random embeddings for STS tasks, almost reach the performance of BERT-derived representations. Our work shows that the representation-shaping techniques significantly improve sentence embeddings extracted from BERT-based and simple baseline models. [ABSTRACT FROM AUTHOR]

This paper proposes a hybrid model to improve Information Content (IC) related metrics of semantic similarity between words, named IC+SP, based on the essential hypothesis that IC and the shortest path are two relatively independent semantic evidences and have approximately equal influences to the semantic similarity metric. The paradigm of IC+SP is to linearly combine the IC-related metric and the shortest path. Meanwhile, a transformation from the semantic similarity of the concepts to that of the words is presented by maximizing every component of IC+SP. 13 improved IC-related metrics based on IC+SP are formed and implemented on the experimental platform HESML Lastra-Díaz (Inf Syst 66:97–118, 2017). Pearson's and Spearman's correlation coefficients on well-accepted benchmarks for the improved metrics compare to those for the original ones to evaluate IC+SP. I introduce the Wilcoxon Signed-Rank Test needing no standard distribution hypothesis, while, this hypothesis is required by T-Test on the sample of small size. T-Test, as well as the Wilcoxon Signed-Rank Test, conduct on the differences of the correlative coefficients for improved and original metrics. It is expected that the improved IC-related metrics could significantly outperform their corresponding original ones, and the experimental results, including the comparisons of mean and maximum of correlation coefficients as well as the p-value and confidence interval of both tests, accomplish the anticipation in the vast majority of cases. [ABSTRACT FROM AUTHOR]

The high dimensionality of text data is a challenging issue that requires efficient methods to reduce vector space and improve classification accuracy. Existing filter-based methods fail to address the redundancy issue, resulting in the selection of irrelevant and redundant features. Information theory-based methods effectively solve this problem but are not practical for large amounts of data due to their high time complexity. The proposed method, termed semantic similarity-aware feature selection and redundancy removal (SS-FSRR), employs joint mutual information between the pairs of semantically related terms and the class label to capture redundant features. It is predicated on the assumption that semantically related terms imply potentially redundant ones, which can significantly reduce execution time by avoiding sequential search strategies. In this work, we use Word2Vec's CBOW model to obtain semantic similarity between terms. The efficiency of the SS-FSRR is compared to six state-of-the-art competitive selection methods for categorical data using two traditional classifiers (SVM and NB) and a robust deep learning model (LSTM) on seven datasets with 10-fold cross-validation, where experimental results show that the SS-FSRR outperforms the other methods on most tested datasets with high stability as measured by the Jaccard's Index. [ABSTRACT FROM AUTHOR]

Background: Bio-ontologies are keys in structuring complex biological information for effective data integration and knowledge representation. Semantic similarity analysis on bio-ontologies quantitatively assesses the degree of similarity between biological concepts based on the semantics encoded in ontologies. It plays an important role in structured and meaningful interpretations and integration of complex data from multiple biological domains. Results: We present simona, a novel R package for semantic similarity analysis on general bio-ontologies. Simona implements infrastructures for ontology analysis by offering efficient data structures, fast ontology traversal methods, and elegant visualizations. Moreover, it provides a robust toolbox supporting over 70 methods for semantic similarity analysis. With simona, we conducted a benchmark against current semantic similarity methods. The results demonstrate methods are clustered based on their mathematical methodologies, thus guiding researchers in the selection of appropriate methods. Additionally, we explored annotation-based versus topology-based methods, revealing that semantic similarities solely based on ontology topology can efficiently reveal semantic similarity structures, facilitating analysis on less-studied organisms and other ontologies. Conclusions: Simona offers a versatile interface and efficient implementation for processing, visualization, and semantic similarity analysis on bio-ontologies. We believe that simona will serve as a robust tool for uncovering relationships and enhancing the interoperability of biological knowledge systems. [ABSTRACT FROM AUTHOR]

With the emerging new technologies based on Artificial Intelligence (AI) for the generation of new and paraphrasing of existing text, the identification of genuinely written text has become an important research undertaking. Past approaches to address this issue, need a significant volume of human-labeled data. Most of the approaches used in literature are either for noisy text or for clean text. Conversations in chats, text in blogs, text messages on cell phones, text exchange on Messengers, etc., are examples of noisy text that may contain misspelled words or incomplete words. The second approach focuses on clean text, which is free from the mentioned characteristics in the noisy text. As research articles do not contain noisy data, we propose a model that focuses on clean text for the identification of paraphrases in research articles. To address the problem of paraphrase detection, this work presents a novel Bidirectional Encoder Representation from Transformers (BERT) based model with fine-tuning. For word representation, Global Vectors (Glove) embeddings and contextualized Embeddings From Language Models (ELMo) are employed in this work. Initially, the model is evaluated without performing preprocessing. Later, the preprocessing step is performed before evaluating the model. Extensive experimentations are performed to evaluate the proposed model utilizing two benchmark datasets, namely, Microsoft Research Paraphrase (MSRP) and Quora Question Pairs (Quora). A comparison of the proposed model is done with four closely related state-of-the-art works. The obtained results show that Fine-tuned BERT using ELMo embeddings with preprocessing produces promising outcomes. Paraphrase identification rates achieved on MSRP and Quora datasets are 86.51% and 94.32%, respectively, which are better than the other competing methods. The proposed solution enables the identification of paraphrased text with a higher accuracy having its application in multiple domains requiring genuinely written documents. [ABSTRACT FROM AUTHOR]

Image forgery detection is essential for forensic and security applications. In local locations where key points (KPs) are few, traditional approaches that rely on them suffer from restrictions. To tackle this problem, this paper suggests an innovative strategy that combines pre-processing, feature extraction, and classification methods. We specifically present EFP-AEHO, or Enhanced Flower Pollination-Adaptive Elephant Herd Optimization, for modifying the Deep Belief Network's (DBN) weights. Our method provides better performance by first converting RGB images into the Lab color space, then applying Eigenvalue Asymmetry (EAS) to extract texture features. Its efficacy is highlighted by experimental findings on the F2000 dataset, which show testing accuracy of 99.4%, TPR of 98%, and TNR of 99.12%. [ABSTRACT FROM AUTHOR]

Aspect-Based Sentiment Analysis (ABSA) for financial review still has some errors which impact the low accuracy of ABSA performance. One of the errors happens because there are implicit aspects and opinions, which are indicated as polysemy terms. For example, the term spread can refer to the aspect term for financial aspect categories and can also refer to the opinion term. Implicit opinion extraction in financial reviews requires in-depth attention because several opinion terms contained in reviews are nouns and not adjectives or verbs that describe explicit opinions (e.g., the review Copper market may get a 2003-style supply shock from Glencore closures). This research proposes ABSA for financial review with implicit aspects and opinions using Semantic similarity and a hybrid approach. We use the FiQA 2018 dataset, which has been classified into four aspect categories: Corporate, Economy, Stock, and Market. First, the dataset is pre-processed. Then, we extract aspect category keywords from Wikipedia using Word2vec. For the aspect categorization method, we use implicit aspect extraction, Semantic similarity, and hybrid BERTBiLSTM to calculate the similarity between extracted aspect terms and aspect category keywords for determining the aspect category. For the ABSA method, we use implicit opinion extraction and hybrid BERT-BiLSTM. The obtained performance result of aspect categorization reaches 91% and the obtained performance result of ABSA reaches 92%. [ABSTRACT FROM AUTHOR]

Mnemonic discrimination of highly similar memory traces is affected in healthy aging via changes in hippocampal pattern separation—i.e., the ability of the hippocampus to orthogonalize highly similar neural inputs. The decline of this process leads to a loss of episodic specificity. Because previous studies have almost exclusively tested mnemonic discrimination of visuospatial stimuli (e.g., objects or scenes), less is known about age-related effects on the episodic specificity of semantically similar traces. To address this gap, we designed a task to assess mnemonic discrimination of verbal stimuli as a function of semantic similarity based on word embeddings. Forty young (Mage = 21.7 years) and 40 old adults (Mage = 69.8 years) first incidentally encoded adjective-noun phrases, then performed a surprise recognition test involving exactly repeated and highly similar lure phrases. We found that increasing semantic similarity negatively affected mnemonic discrimination in both age groups, and that compared to young adults, older adults showed worse discrimination at medium levels of semantic similarity. These results indicate that episodic specificity of semantically similar memory traces is affected in aging via less efficient mnemonic operations and strengthen the notion that mnemonic discrimination is a modality-independent process supporting memory specificity across representational domains. [ABSTRACT FROM AUTHOR]

Pavement cracks are crucial indicators for assessing the structural health of asphalt roads. Existing automated crack detection models depend on large quantities of precisely annotated crack sample data. The irregular morphology of cracks makes manual annotation time-consuming and costly, thereby posing challenges to the practical application of these models. This study proposes a pavement crack image detection method integrating fractal dimension analysis and semi-supervised learning. It identifies the self-similarity characteristics within the crack regions by analyzing pavement crack images and using fractal dimensions to preliminarily determine the candidate crack regions. The Crack Similarity Learning Network (CrackSL-Net) is then employed to learn the semantic similarity of crack image regions. Semi-supervised learning facilitates automatic crack detection by combining a small amount of labeled data with a large volume of unlabeled image data. Comparative experiments are conducted on two public pavement crack datasets against the HED, U-Net, and RCF models to comprehensively evaluate the performance of the proposed method. The results indicate that, with a 50% annotation ratio, the proposed method achieves high-precision crack detection, with an intersection over union (IoU) exceeding 0.84, which is close to that of U-Net. Visual analysis of the detection results confirms the method's effectiveness in identifying cracks in complex environments. [ABSTRACT FROM AUTHOR]

Contrastive learning utilizes data augmentation to guide network training. This approach has attracted considerable attention for clustering, object detection, and image segmentation. However, previous studies have ignored the impact of false-negative pairs, resulting in the dissimilarity of the semantic representations of the same cluster. Some researchers have attempted to address this problem; however, only considering the image level has provided unsatisfactory results. To this end, we propose a novel feature extraction algorithm suitable for clustering, combining core points and semantic similarity at the cluster level to restructure positive and negative pairs. Specifically, the core points consisting of the n-nearest neighbors of the cluster center are considered the semantic sample relations of the cluster. This information is explored to reconstruct semantic positive and negative pairs to maximize intra-cluster similarity and inter-cluster variability. More accurate cluster centers offer a sub-optimal initialization for updating the feature model and clustering assignment, which is optimized by the expectation-maximization framework. Extensive experiments conducted on six benchmark datasets show promising clustering performances with relatively few training epochs. The proposed method outperforms the best baseline by 4 % (1.5 % ) on CIFAR-100 (CIFAR-10). The CPCS code is open-sourced at https://github.com/Cappuccino-Sugar/CPCS. [ABSTRACT FROM AUTHOR]

Abstract Background Bio-ontologies are keys in structuring complex biological information for effective data integration and knowledge representation. Semantic similarity analysis on bio-ontologies quantitatively assesses the degree of similarity between biological concepts based on the semantics encoded in ontologies. It plays an important role in structured and meaningful interpretations and integration of complex data from multiple biological domains. Results We present simona, a novel R package for semantic similarity analysis on general bio-ontologies. Simona implements infrastructures for ontology analysis by offering efficient data structures, fast ontology traversal methods, and elegant visualizations. Moreover, it provides a robust toolbox supporting over 70 methods for semantic similarity analysis. With simona, we conducted a benchmark against current semantic similarity methods. The results demonstrate methods are clustered based on their mathematical methodologies, thus guiding researchers in the selection of appropriate methods. Additionally, we explored annotation-based versus topology-based methods, revealing that semantic similarities solely based on ontology topology can efficiently reveal semantic similarity structures, facilitating analysis on less-studied organisms and other ontologies. Conclusions Simona offers a versatile interface and efficient implementation for processing, visualization, and semantic similarity analysis on bio-ontologies. We believe that simona will serve as a robust tool for uncovering relationships and enhancing the interoperability of biological knowledge systems.

Abstract Mnemonic discrimination of highly similar memory traces is affected in healthy aging via changes in hippocampal pattern separation—i.e., the ability of the hippocampus to orthogonalize highly similar neural inputs. The decline of this process leads to a loss of episodic specificity. Because previous studies have almost exclusively tested mnemonic discrimination of visuospatial stimuli (e.g., objects or scenes), less is known about age-related effects on the episodic specificity of semantically similar traces. To address this gap, we designed a task to assess mnemonic discrimination of verbal stimuli as a function of semantic similarity based on word embeddings. Forty young (Mage = 21.7 years) and 40 old adults (Mage = 69.8 years) first incidentally encoded adjective-noun phrases, then performed a surprise recognition test involving exactly repeated and highly similar lure phrases. We found that increasing semantic similarity negatively affected mnemonic discrimination in both age groups, and that compared to young adults, older adults showed worse discrimination at medium levels of semantic similarity. These results indicate that episodic specificity of semantically similar memory traces is affected in aging via less efficient mnemonic operations and strengthen the notion that mnemonic discrimination is a modality-independent process supporting memory specificity across representational domains.

Semantic similarity has been gaining traction in the field of natural language processing. It is a measure of how similar two pieces of text are in terms of their meaning. It can be used to improve search engine results. We propose a deep learning-based approach to build a semantic similarity search engine for movies based on a movie summary. Filmmakers can gain insight into audience preferences and trends, allowing them to create more engaging and successful films. The dataset used in this study was gathered from internet movie database (IMDb), it includes movie summaries along with their corresponding name movies. The test dataset was generated using ChatGPT to be very close to human input. The universal sentence encoder (USE) model presented promising results in semantic similarity, the model results show that for the top 5 similar movies, the model returned 176 out of 300 movies (58.6%). For the top 10 similar movies, the model returned 211 out of 300 movies (70.3%). Additionally, for the top 15 similar movies, the model returned 229 out of 300 movies (76.3%). And, for the top 20 similar movies, the model returned 249 of 300 movies (83%). This method can be applied to movie recommendation systems or to organize films in a collection automatically. [ABSTRACT FROM AUTHOR]

Homonym and polysemy features can influence some errors in translation from a source language to another target language, for example, from Indonesian to English. A lemma or a morphology factor can cause the configuration of Indonesian homonym features. For example, the word beruang can mean an animal beruang (bear) and can mean a verb alternation ber+uang (has/have money). The Indonesian polysemy feature can also impact an error in the translation process because it can have a literal meaning and a symbolic meaning. For example, the terms bunga melati (jasmine flower) and bunga hati (lover), where bunga does not only mean flower. Therefore, the development machine translation (MT) method needs to capture homonym and polysemy features in the form of a word or a phrase. This research proposes homonym and polysemy approaches with morphology extraction in weighting terms for Indonesian to English MT. First, this research uses morphology extraction to detect sentences that contain prefixes, lemma, and suffixes. Then, the word similarity measurement functions to extract homonym and polysemy in the form of uni-gram and bi-gram using bidirectional encoder representations from transformers (BERT) embedding, named entity recognition (NER), synonym-based term expansion, and semantic similarity. This research uses neural machine translation for the translation process. [ABSTRACT FROM AUTHOR]

The different types of implicit reviews should be well understood so that the developed extraction technique can solve all problems in implicit reviews and produce precise terms of aspects and opinions. We propose an aspectbased sentiment analysis (ABSA) method with natural language understanding for implicit reviews based on sentence and word structure. We built a text extraction method using a machine learning algorithm rule with a deep understanding of different types of sentences and words. Furthermore, the aspect category of each review is determined by measuring the word similarity between the aspect terms contained in each review and aspect keywords extracted from Wikipedia. Bidirectional encoder representations from transformers (BERT) embedding and semantic similarity are used to measure the word similarity value. Moreover, the proposed ABSA method uses BERT, a hybrid lexicon, and manual weighting of opinion terms. The purpose of the hybrid lexicon and the manual weighting of opinion terms is to update the existing lexicon and solve the problem of weighting words and phrases of opinion terms. The evaluation results were very good, with average F1-scores of 93.84% for aspect categorization and 92.42% for ABSA. [ABSTRACT FROM AUTHOR]

Abstract This paper proposes a hybrid model to improve Information Content (IC) related metrics of semantic similarity between words, named IC+SP, based on the essential hypothesis that IC and the shortest path are two relatively independent semantic evidences and have approximately equal influences to the semantic similarity metric. The paradigm of IC+SP is to linearly combine the IC-related metric and the shortest path. Meanwhile, a transformation from the semantic similarity of the concepts to that of the words is presented by maximizing every component of IC+SP. 13 improved IC-related metrics based on IC+SP are formed and implemented on the experimental platform HESML Lastra-Díaz (Inf Syst 66:97–118, 2017). Pearson’s and Spearman’s correlation coefficients on well-accepted benchmarks for the improved metrics compare to those for the original ones to evaluate IC+SP. I introduce the Wilcoxon Signed-Rank Test needing no standard distribution hypothesis, while, this hypothesis is required by T-Test on the sample of small size. T-Test, as well as the Wilcoxon Signed-Rank Test, conduct on the differences of the correlative coefficients for improved and original metrics. It is expected that the improved IC-related metrics could significantly outperform their corresponding original ones, and the experimental results, including the comparisons of mean and maximum of correlation coefficients as well as the p-value and confidence interval of both tests, accomplish the anticipation in the vast majority of cases.

Semantic Textual Similarity (STS) aims to assess the semantic similarity between two pieces of text. As a challenging task in natural language processing, various approaches for STS in high-resource languages, such as English, have been proposed. In this paper, we are concerned with STS in low resource languages such as Arabic. A baseline approach for STS is based on vector embedding of the input text and application of similarity metric on the embedding space. In this contribution, we propose a cross-encoder neural network (Cross-BERT-GRU) to handle semantic similarity of Arabic sentences that benefits from both the strong contextual understanding of BERT and the sequential modeling capabilities of GRU. The architecture begins by inputting the BERT word embeddings for each word into a GRU cell to model long-term dependencies. Then, max pooling and average pooling are applied to the hidden outputs of the GRU cell, serving as the sentence -pair encoder. Finally, a softmax layer is utilized to predict the degree of similarity. The experiment results show a Spearman correlation coefficient of around 0.9 and that Cross-BERT-GRU outperforms the other BERT models in predicting the semantic textual similarity of Arabic sentences. The experimentation results also indicate that the performance improves by integrating data augmentation techniques.

Advances in computational language models increasingly enable adaptive support for self‐regulated learning (SRL) in digital learning environments (DLEs; eg, via automated feedback). However, the accuracy of those models is a common concern for educational stakeholders (eg, policymakers, researchers, teachers and learners themselves). We compared the accuracy of four Dutch language models (ie, spaCy medium, spaCy large, FastText and ConceptNet NumberBatch) in the context of secondary school students' learning of causal relations from expository texts, scaffolded by causal diagram completion. Since machine learning relies on human‐labelled data for the best results, we used a dataset with 10,193 students' causal diagram answers, compiled over a decade of research using a diagram completion intervention to enhance students' monitoring of their text comprehension. The language models were used in combination with four popular machine learning classifiers (ie, logistic regression, random forests, support vector machine and neural networks) to evaluate their performance on automatically scoring students' causal diagrams in terms of the correctness of events and their sequence (ie, the causal structure). Five performance metrics were studied, namely accuracy, precision, recall, F1 and the area under the curve of the receiver operating characteristic (ROC‐AUC). The spaCy medium model combined with the neural network classifier achieved the best performance for the correctness of causal events in four of the five metrics, while the ConceptNet NumberBatch model worked best for the correctness of the causal sequence. These evaluation results provide a criterion for model adoption to adaptively support SRL of causal relations in DLEs. Practitioner notesWhat is already known about this topic Accurate monitoring is a prerequisite for effective self‐regulation.Students struggle to accurately monitor their comprehension of causal relations in texts.Completing causal diagrams improves students' monitoring accuracy, but there is room for further improvement.Automatic scoring could be used to provide adaptive support during diagramming.What this paper adds Comparison of four Dutch word vector models combined with four machine learning classifiers for the automatic scoring of students' causal diagrams.Five performance metrics to evaluate the above solutions.Evaluation of the word vector models for estimating the semantic similarity between student and model answers.Implications for practice and/or policy High‐quality word vector models could (em)power adaptive support during causal diagramming via automatic scoring.The evaluated solutions can be embedded in digital learning environments (DLEs).Criteria for model adoption to adaptively support SRL of causal relations in DLEs.The increased saliency of (in)correct answers via automatic scoring might help to improve students' monitoring accuracy. [ABSTRACT FROM AUTHOR]

Mental disorders significantly impact daily life and are among the leading causes of suicide. Despite numerous studies on detecting mental disorders on social media, the focus has primarily been on identifying the presence or absence of indications in posts, with most studies concentrating solely on one specific mental disorder, particularly depression. There is a lack of comprehensive analysis of the detection results. Therefore, this study analyzes mental disorders in more detail by applying detection and sentiment analysis based on five aspects, namely ADHD (attention-deficit hyperactivity disorder), anxiety, bipolar, depression, and PTSD (post-traumatic stress disorder). The detection process utilizes bidirectional encoder representations from transformers (BERT) embedding and the bidirectional gated recurrent unit (BiGRU) model. Subsequently, aspect categorization employs semantic similarity, which assesses the resemblance between terms generated from hidden topic extraction via non-negative matrix factorization (NMF) and keywords linked to the five mental disorder aspects, extracted using a combination of term extraction methods. Additionally, sentiment classification leverages BERT embedding and the BiGRU model. The proposed method successfully identifies mental disorders, categorizes aspects, and classifies sentiment accurately. Optimal performance is achieved in mental disorders detection (0.9009) using BERT embedding + BiGRU, aspect categorization (0.8507) employing semantic similarity + BiGRU, and sentiment classification (0.8717) through BERT embedding + BiGRU. The analysis results unveil that texts related to mental disorders often convey negative sentiments, with the depression aspect exhibiting higher percentages of negative sentiment compared to other mental disorder aspects. [ABSTRACT FROM AUTHOR]

This paper aims to highlight the latest and most widely used artificial intelligence methods in text similarity prediction. The study focuses on articles published from 2010 to 2022 in the field of text similarity prediction and algorithms applied in this field. In this paper, systematic scientific comparisons have been made between existing approaches for predicting text similarity to answer the question raised in this study about the most used and accurate approach. Through previous studies and the comparisons made in this paper, the Ant Colony Optimization Algorithm (ACO) approach was found to be the most frequently used in text similarity prediction and solving scheduling problems. [ABSTRACT FROM AUTHOR]

Among the possible solutions for automatic lexical disambiguation in natural language processing tasks, we find methods based on machine learning algorithms, semantic relatedness, and semantic similarity measures. While machine learning methods use endogenous sources of knowledge, semantic relatedness and similarity measures resort to exogenous sources of knowledge, such as definitions from lexicographic resources or lexical meaning relations from ontologies or thesauri, which offer a conceptual hierarchy. In this work, we present and analyze the different types of methods for automatic lexical disambiguation divided into four groups: based on machine learning algorithms, based on semantic relatedness measures, based on semantic similarity measures, and based on hybrid measures. We postulate that the advantage of methods based on relationship and similarity measures lies in the fact that their results are derived from statistical efficiency and linguistic knowledge found in the parameters that make up each of the measures used. [ABSTRACT FROM AUTHOR]

Most trajectory compression methods primarily focus on geometric similarity between compressed and original trajectories, lacking explainability of compression results due to ignoring semantic information. This paper proposes a spatio-temporal semantic constrained trajectory compression method. It constructs a new trajectory distance measurement model integrating both semantic and spatio-temporal features. This model quantifies semantic features using information entropy and measures spatio-temporal features with synchronous Euclidean distance. The compression principle is to retain feature points with maximum spatio-temporal semantic distance from the original trajectory until the compression rate is satisfied. Experimental results show these methods closely resemble each other in maintaining geometric similarity of trajectories, but our method significantly outperforms DP, TD-TR, and CascadeSync methods in preserving semantic similarity of trajectories. This indicates that our method considers both geometric and semantic features during compression, resulting in the compressed trajectory becoming more interpretable. [ABSTRACT FROM AUTHOR]

Streaming services are increasingly leveraging Artificial Intelligence (AI) technologies for improved content cataloging, user experiences in content discovery, and personalization. A significant challenge in this domain is the automated assignment of microgenres to movies. This study introduces and evaluates approaches based on clustering, topic modeling, and word embedding to address this task. The evaluation employs a preprocessed dataset containing movie-related data—title tags, synopses, genres, and reviews—alongside a predefined microgenre list. Comparisons of three activation functions (binary step, ramp, and sigmoid) gauge their effectiveness in augmenting microgenre tags. Results demonstrate the superiority of the word embedding approach over clustering and topic modeling in terms of mean accuracy. Even more, the word embedding approach stands as the sole fully automated solution. Analysis indicates that incorporating review-based tags introduces noise and undermines accuracy. Besides, the word embedding approach yields optimal outcomes using the sigmoid function, effectively doubling assigned tags while maintaining matching quality. This sheds light on the potential of word embedding methods within the movie domain. [ABSTRACT FROM AUTHOR]

High-dimensional numerical vectors are widely used in machine learning for searching and indexing data. However, it is often difficult for users to interpret their meaning. To address this, we introduce a novel approach that transforms dense vectors into human-readable lexical representations using a percentile-based mapping approach. The essence of the approach is a mapping of words from a predefined/custom lexicon to vectors based on their relative local magnitudes. This way, it enables intuitive visualization of the semantic similarities and differences between complex data points and allows for domain-specific interpretability. It provides an easy way to deduplicate dense vectors (even near-duplicates) and can generate locality-aware hash-like representations, which can be used for efficient indexing and retrieval in various applications. The approach has also been implemented in an open-source library called LangVec. The paper provides examples on LangVec usage and highlights the key applications, including semantic search, recommendation systems, and clustering of numerical data into a human-readable format. [ABSTRACT FROM AUTHOR]

Modern natural language processing (NLP) methods provide ways to objectively quantify language disturbances for potential use in diagnostic classification. We performed computerized language analysis in speech samples of 82 Turkish-speaking subjects, including 44 patients with schizophrenia spectrum disorders (SSD) and 38 healthy controls (HC). Exploratory analysis of speech samples involved 16 sentence-level semantic similarity features using SBERT (Sentence Bidirectional Encoder Representation from Text) as well as 8 generic and 8 part-of-speech (POS) features. The random forest classifier using SBERT-derived semantic similarity features achieved a mean accuracy of 85.6 % for the classification of SSD and HC. When semantic similarity features were combined with generic and POS features, the classifier's mean accuracy reached to 86.8 %. Our analysis reflected increased sentence-level semantic similarity scores in SSD. Generic and POS analyses revealed an increase in the use of verbs, proper nouns and pronouns in SSD while our results showed a decrease in the utilization of conjunctions, determiners, and both average and maximum sentence length in SSD compared to HC. Quantitative language features were correlated with the expressive deficit domain of BNSS (Brief Negative Symptom Scale) as well as with the duration of illness. These findings from Turkish-speaking interviews contribute to the growing evidence-based NLP-derived assessments in non-English-speaking patients. [ABSTRACT FROM AUTHOR]