Search

To identify machine and human, Completely Automated Public Turing test to tell Computers and Humans Apart (CAPTCHA) is increasingly used in many web applications. The classical English and digital characters based CAPTCHAs are recognized with high accuracy. Due to the complication of Chinese characters which greatly enhance the difficulty of automatic recognition, an increasing number of Chinese web sites use Chinese Character CAPTCHAs. To recognize Chinese Character CAPTCHAs, we propose a Convolution Neural Network (CNN) based approach to learn strokes, radicals and character features of Chinese characters, and prove that our network structure is superior to LENET-5 in this task. Furthermore, we formulate the relation among accuracy, the number of training samples and iterations, which is used to estimate the performance of our approach. Firstly, this approach greatly improves the recognition accuracy of Chinese Character CAPTCHAs with distortion, rotation and background noise. Our experiments results show that this approach achieves over 95% accuracy for single Chinese character and 84% accuracy for three types of Chinese Character CAPTCHAs with four Chinese characters. Secondly, our experiment results and theoretical analysis show that the accuracy of recognition has the exponential relationship with the product of the number of training samples and iterations in the condition of enough and representative training samples. Therefore, we can estimate the training time for a certain accuracy. Finally, we certify that our approach is superior to the most famous Chinese Optical Character Recognition (OCR) software, Hanvon, in Chinese Character CAPTCHAs recognition. [ABSTRACT FROM AUTHOR]

Understanding the sentiments of users from cross media contents which contain texts and images is an important task for many social network applications. However, due to the semantic gap between cross media features and sentiments, machine learning methods need a lot of human labeled samples. Furthermore, for each kind of media content, it is necessary to constantly add a lot of new human labeled samples because of new expressions of sentiments. Fortunately, there are some emotion signals, like emoticons, which denote users’ emotions in cross media contents. In order to use these weakly labels to build a unified multi-modality sentiment learning framework, we propose an Explicit Emotion Signal (EES) based multi-modality sentiment learning approach which uses huge number of weakly labeled samples in sentiment learning. There are three advantages in our approach. Firstly, only a few human labeled samples are needed to reach the same performance which can be obtained by the traditional machine learning based sentiment prediction approaches. Secondly, this approach is flexible and can easily combine text and vision based sentiment learning through deep neural networks. Thirdly, because a lot of weakly labeled samples can be used in EES, trained model is more robust in different domain transfer. In this paper, firstly, we investigate the correlation between sentiments and emoticons and choose emoticons as the Explicit Emotion Signals in our approach; secondly, we build a two stages multi-modality sentiment learning framework based on Explicit Emotion Signals. Our experiment results show that our approach not only achieves the best performance but also only needs 3% and 43% training samples to obtain the same performance of Visual Geometry Group (VGG) model and Long Short-Term Memory (LSTM) model in images and texts, respectively. [ABSTRACT FROM AUTHOR]

With the development of social media, an increasing number of people use short videos in social media applications to express their opinions and sentiments. However, sentiment detection of short videos is a very challenging task because of the semantic gap problem and sequence based sentiment understanding problem. In this context, we propose a SentiPair Sequence based GIF video sentiment detection approach with two contributions. First, we propose a Synset Forest method to extract sentiment related semantic concepts from WordNet to build a robust SentiPair label set. This approach considers the semantic gap between label words and selects a robust label subset which is related to sentiment. Secondly, we propose a SentiPair Sequence based GIF video sentiment detection approach that learns the semantic sequence to understand the sentiment from GIF videos. Our experiment results on GSO-2016 (GIF Sentiment Ontology) data show that our approach not only outperforms four state-of-the-art classification methods but also shows better performance than the state-of-the-art middle level sentiment ontology features, Adjective Noun Pairs (ANPs). [ABSTRACT FROM AUTHOR]

More and more people use blogs to write down their ideas, opinions and individual thoughts. Mining blogs will obtain useful information, which can support business policy and decision-making, especially in analyzing product popularity. In this study, we mine the features of persons as implicit relations between the content of the bloggers’ posts and the bloggers. These kinds of implicit relations, which are also semantic relations, is called the Blogger Role. To mine this semantic relation, Wordnet is used to extract the Blogger Role and the features of Blogger Role. To get an appropriate Blogger Role, we cluster all bloggers’ posts and use the clustering result to revise the Blogger Role obtained by single document analysis. To get more relevant retrieval results, we combine this implicit relation with the classical retrieval model by the Blogger Role-based model. The combination is performed by the explicit model and implicit model. Results of experiments on TREC corpus show that Blogger Role reveals bloggers’ characters and mining Blogger Roles is useful in analyzing the popularity of products. [ABSTRACT FROM AUTHOR]

With the increasing number of vehicles in China, traffic condition analysis is of great significance to urban planning and public administration. However, the state-of-the-art traffic condition analysis approaches mainly rely on sensors, which are high cost and limited coverage. To solve these problems, we propose a semi-supervised learning method which uses the social network data instead and analyzes the traffic condition based on users’ sentiment in Chinese Microblog. This approach is a low-cost crowdsourcing solution. Firstly, we train the gated recurrent unit (GRU) model and generative adversarial networks to estimate the sentiment of Microblog with traffic information. Secondly, we calculate the Traffic Sentiment Index to predict whether traffic jams happen or not. In order to reduce the data annotated by manpower, we propose a new idea to employ the conditional generative adversarial networks to generate robust features which are used as a supplement to the training set of GRU. Finally compared with the GRU model trained by solely the manual annotation data, our method improves the classification accuracy by 3.79%. Furthermore, by using the Traffic Sentiment Index, we build a traffic condition analysis system and predict the time and roads of traffic jams in 4 Chinese cities. The predication experiment shows similar results with Baidu map which uses a lot of mobile phones as sensors, and proves the low-cost characteristic and performance of our approach. [ABSTRACT FROM AUTHOR]

Since classical public sentiment analysis systems for microblog are based on the text sentiment analysis, it is difficult to determine the sentiment of short text without clear sentiment words in microblog posts. Fortunately, a lot of microblog posts contain images which also represent users' sentiment. To fully understand users' sentiment, we propose a cross-media public sentiment analysis system for microblog. The best advantage of this novel system is the unified cross-media public sentiment analysis framework which fuses the text sentiment and image sentiment not only from sentiment results, but also from sentiment ontology. To enhance presentation effects, this system presents sentiment results from macroscopic view and microscopic view which details the sentiment results in region, topic, microblog content and user diffusion. In our knowledge, this is the first unified cross-media public sentiment analysis system. [ABSTRACT FROM AUTHOR]

Semi-supervised learning has garnered significant interest as a method to alleviate the burden of data annotation. Recently, semi-supervised medical image segmentation has garnered significant interest that can alleviate the burden of densely annotated data. Substantial advancements have been achieved by integrating consistency-regularization and pseudo-labeling techniques. The quality of the pseudo-labels is crucial in this regard. Unreliable pseudo-labeling can result in the introduction of noise, leading the model to converge to suboptimal solutions. To address this issue, we propose learning from reliable pseudo-labels. In this paper, we tackle two critical questions in learning from reliable pseudo-labels: which pseudo-labels are reliable and how reliable are they? Specifically, we conduct a comparative analysis of two subnetworks to address both challenges. Initially, we compare the prediction confidence of the two subnetworks. A higher confidence score indicates a more reliable pseudo-label. Subsequently, we utilize intra-class similarity to assess the reliability of the pseudo-labels to address the second challenge. The greater the intra-class similarity of the predicted classes, the more reliable the pseudo-label. The subnetwork selectively incorporates knowledge imparted by the other subnetwork model, contingent on the reliability of the pseudo labels. By reducing the introduction of noise from unreliable pseudo-labels, we are able to improve the performance of segmentation. To demonstrate the superiority of our approach, we conducted an extensive set of experiments on three datasets: Left Atrium, Pancreas-CT and Brats-2019. The experimental results demonstrate that our approach achieves state-of-the-art performance. Code is available at: https://github.com/Jiawei0o0/mutual-learning-with-reliable-pseudo-labels. • We propose a mutual learning method for semi-supervised medical image segmentation. • We suggest using a mutual comparison approach to provide reliable pseudo labels. • We introduce intra-class consistency to evaluate the reliability of pseudo labels. • We achieve state-of-the-art performance on three benchmark datasets. [ABSTRACT FROM AUTHOR]

Fuzzy cerebellar model articulation controller (FCMAC) networks with excellent nonlinear appropriation ability and simple implementation are used to solve complex uncertainties problems in engineering applications. Both online and off-line learning algorithm of FCMAC networks usually applies the gradient-descent-type methods. However, such gradient-descent methods lead to the high possibility to converging into local minima. To cope with the local minimum problem, this paper alternatively proposes to apply harmony search algorithm to find optimal network parameters, so as to achieve better performances of FCMAC. The harmony search algorithm optimizes not only FCMAC network’s weight variables, but also optimizes network receptive field’s center position and standard deviation parameters. In order to obtain an optimal network, the weight values, center positions, and standard deviations are transformed to three data strings that can be processed by harmony search algorithm. In particular, the self-adaptive global best harmony search algorithm (SGHS) is used to search optimal parameter combinations of FCMAC within solution domains. The network’s performances are verified by approximating six nonlinear formulae. In order to compare the performances of the FCMAC networks optimized by the SGHS algorithm, a back-propagation trained network and another harmony search variant optimized networks are also tested in this work. The experimental results show that the networks optimized by SGHS perform the faster convergence speed and better accuracy. [ABSTRACT FROM AUTHOR]