Jiajun Cheng

Exploring sentiment parsing of microblogging texts for opinion polling on chinese public figures

Microblogging websites such as twitter and Sina Weibo have attracted many users to share their experiences and express their opinions on a variety of topics, making them ideal platforms on which to conduct electronic opinion polls on products, services and public figures. However, conventional sentiment analysis methods for microblogging messages may not meet the demands of opinion polls for public figures. Therefore, in this study, we focus mainly on the problem of sentiment analysis for opinion polling on Chinese public figures. We propose a sentiment parsing-based architecture, which represents and labels opinion targets and their corresponding sentiments jointly to avoid the mismatching of them, for opinion poll of public figures using microblogs. Furthermore, we formulate sentiment parsing of microblogging sentences as a sequence labeling problem and adapt different Recurrent Neural Network (RNN) models to train and infer the model. Our experimental results demonstrate that the proposed sentiment parsing-based methods achieve better performance than conventional sentiment score-based methods in opinion polling on public figures using microblogs.

Aspect-level Sentiment Classification with HEAT (HiErarchical ATtention) Network

Aspect-level sentiment classification is a fine-grained sentiment analysis task, which aims to predict the sentiment of a text in different aspects. One key point of this task is to allocate the appropriate sentiment words for the given aspect.Recent work exploits attention neural networks to allocate sentiment words and achieves the state-of-the-art performance. However, the prior work only attends to the sentiment information and ignores the aspect-related information in the text, which may cause mismatching between the sentiment words and the aspects when an unrelated sentiment word is semantically meaningful for the given aspect. To solve this problem, we propose a HiErarchical ATtention (HEAT) network for aspect-level sentiment classification. The HEAT network contains a hierarchical attention module, consisting of aspect attention and sentiment attention. The aspect attention extracts the aspect-related information to guide the sentiment attention to better allocate aspect-specific sentiment words of the text. Moreover, the HEAT network supports to extract the aspect terms together with aspect-level sentiment classification by introducing the Bernoulli attention mechanism. To verify the proposed method, we conduct experiments on restaurant and laptop review data sets from SemEval at both the sentence level and the review level. The experimental results show that our model better allocates appropriate sentiment expressions for a given aspect benefiting from the guidance of aspect terms. Moreover, our method achieves better performance on aspect-level sentiment classification than state-of-the-art models.

Predicting Social Unrest Events with Hidden Markov Models Using GDELT

Proactive handling of social unrest events which are common happenings in both democracies and authoritarian regimes requires that the risk of upcoming social unrest event is continuously assessed. Most existing approaches comparatively pay little attention to considering the event development stages. In this paper, we use autocoded events dataset GDELT (Global Data on Events, Location, and Tone) to build a Hidden Markov Models (HMMs) based framework to predict indicators associated with country instability. The framework utilizes the temporal burst patterns in GDELT event streams to uncover the underlying event development mechanics and formulates the social unrest event prediction as a sequence classification problem based on Bayes decision. Extensive experiments with data from five countries in Southeast Asia demonstrate the effectiveness of this framework, which outperforms the logistic regression method by 7% to 27% and the baseline method 34% to 62% for various countries.

CNN-Based Sequence Labeling for Fine-Grained Opinion Mining of Microblogs

Opinion mining on microblogs is of significance because microblogging websites have attracted many users to share their experiences and express their opinions on a variety of topics. However, conventional opinion mining methods focus mainly on sentiment of texts and ignore opinion target. This paper focuses on a fine-grained opinion mining task that jointly extract opinion target and corresponding sentiment by sequence labeling. We propose a convolutional neural network (CNN)-based sequence labeling method and apply it to fine-grained opinion mining of microblogs. We empirically evaluated neural networks with different filter length and depth and analyzed the boundary of contextual feature extraction for opinion mining of microblogs. The experimental results demonstrate that the proposed CNN-based methods are better than RNN-based methods in both effectiveness and efficiency.

Embedding-Based Feature Extraction Methods for Chinese Sentiment Classification

Sentiment analysis, also known as opinion mining, seeks to figure out points of view from documents. Sentiment classification is a specific task of sentiment analysis that divides documents into positive and negative sentiment polarities according to the attitudes expressed. Feature extraction is a significant part of sentiment classification. Traditional feature extraction methods mine statistical information in documents but neglect semantic relationships between words, while some embedding methods successfully capture semantics but have difficulty in distinguishing sentiment polarities. In this paper, we propose three different sentiment specific models that take advantages of the statistical information in a document as well as the semantic relationships between words. Our models shed more light on the different roles of words in documents, assigning them different weights. Experimental results on three Chinese datasets illustrate that, in general, our models are superior to other models and provide state-of-the-art performance. Our models are efficient and have high stability.

Deep convRNN for sentiment parsing of Chinese microblogging texts

Recurrent Neural Networks (RNNs) are naturally applicable to sequential processing and have achieved outstanding performance in analyzing natural language. However, RNN-based sequence labeling methods may encounter some problems, such as word ambiguous and low-fidelity of word segmentation, in sentiment parsing of Chinese microblogging texts because it cannot well grasp local contextual information of words. Therefore, in this work, we propose a novel neural network architecture, named convRNN, for sentiment parsing of Chinese texts. The convRNN combines Convolutional Neural Network (CNN) and RNN to capture the local contextual feature and global sequence feature of words in a sentence. Experimental results demonstrate that extracting local contextual features of words with CNN improves the performance of RNN models. Furthermore, deep convRNNs achieve better performance than shallow models and outperform the RNN-based method substantially.

Finding Influential Papers in Citation Networks

The citation network is the social network made up of papers and their citation relationships. A key task of the citation network is to find influential papers in the network. Traditional properties such as centralities can not reflect the influence of papers comprehensive, since it does not take the authority of the paper and the transfer effect into account. Other algorithms such as PageRank and HITS overcome those shortcomings. However, both of them involve matrix multiplication and repeated iterative process, which is less-effective. Comparing with another often mentioned network, the coauthor networks, we noticed that the citation network is a Directed Acyclic Graph(DAG) and it has a transitive relation. Making the most of properties of the citation network, we draw on the thought of topological sorting and design a more effective algorithm, whose time complexity is linear with the number of vertices and edges, which is O(V+E). In addition, we illustrate that the algorithm we proposed is stable and effective. We also apply our algorithm to another DAG task and results show that our algorithm has great scalability.