Search

This article presents analysis of data from scientific articles and patents to identify the evolving trends and underlying topics in research on humanoid robots. We used topic modelling based on latent Dirichlet allocation analysis to identify underlying topics in sub-areas in the field. We also used social network analysis to measure the centrality indices of publication keywords to detect important and influential sub-areas and used co-occurrence analysis of keywords to visualise relationships among subfields. The research result is useful to identify evolving topics and areas of current focus in the field of humanoid technology. The results contribute to identify valuable research patterns from publications and to increase understanding of the hidden knowledge themes that are revealed by patents. [ABSTRACT FROM AUTHOR]

Large quantities of textual posts are increasingly generated in course discussion forums, and the accumulation of these data greatly increases the cognitive loads on online participants. It is imperative for them to automatically identify the potential semantic information derived from these textual discourse interactions. Moreover, existing topic models can discover the latent topics or sentimental polarities from textual data, but these models typically ignore the interactive ways of discussing topics, thus making it difficult to further construct topics' semantic space from the perspective of document generation. To solve this issue, we proposed a joint sentiment and behaviour topic model called SBTM, which was an unsupervised approach for automatic analysis of learners' discussed posts. The results demonstrated that SBTM was quantitatively effective on both model generalisation and topic exploration, and rich topic content was qualitatively characterised. Furthermore, the model can be potentially employed in some practical applications, such as information summarisation and behaviour-oriented personalised recommendation. [ABSTRACT FROM AUTHOR]

Modelling short text is challenging due to the small number of word co-occurrence and insufficient semantic information that affects downstream Natural Language Processing (NLP) tasks, for example, text classification. Gathering information from external sources is expensive and may increase noise. For efficient short text classification without depending on external knowledge sources, we propose Expressive Short text Classification (EStC). EStC consists of a novel document context-aware semantically enriched topic model called the Short text Topic Model (StTM) that captures words, topics and documents semantics in a joint learning framework. In StTM, the probability of predicting a context word involves the topic distribution of word embeddings and the document vector as the global context, which obtains by weighted averaging of word embeddings on the fly simultaneously with the topic distribution of words without requiring an additional inference method for the document embedding. EStC represents documents in an expressive (number of topics × number of word embedding features) embedding space and uses a linear support vector machine (SVM) classifier for their classification. Experimental results demonstrate that EStC outperforms many state-of-the-art language models in short text classification using several publicly available short text data sets. [ABSTRACT FROM AUTHOR]

In recent years, we have been faced with a series of natural disasters causing a tremendous amount of financial, environmental and human losses. The unpredictable nature of natural disasters behaviour makes it hard to have a comprehensive situational awareness (SA) to support disaster management. Using opinion surveys is a traditional approach to analyse public concerns during natural disasters; however, this approach is limited, expensive and time-consuming. Luckily, the advent of social media has provided scholars with an alternative means of analysing public concerns. Social media enable users (people) to freely communicate their opinions and disperse information regarding current events including natural disasters. This research emphasises the value of social media analysis and proposes an analytical framework: Twitter Situational Awareness (TwiSA). This framework uses text mining methods including sentiment analysis and topic modelling to create a better SA for disaster preparedness, response and recovery. TwiSA has also effectively deployed on a large number of tweets and tracks the negative concerns of people during the 2015 South Carolina flood. [ABSTRACT FROM AUTHOR]

Understandings of racism are produced and circulated in contemporary networked media forms, contributing new opportunities to both extend and challenge racializing discourses, images and frameworks. This study investigated how the concept of racism is used in Finnish public debate by employing a computational text analysis technique to derive topics related to racism from a large corpus of news media content and online discussion forum comments. Our findings show that discourses about racism are different in legacy media and online platform regarding both their prominence and framings. While social media produce various discourses of 'reverse racism', news media connects racism to historical and international contexts. We conclude that what racism is understood to be is not only an intensely political process but also one shaped by the type of media platform, specificities of Finnish language and national ideological battles. [ABSTRACT FROM AUTHOR]

Consumers usually do not know the complicated links between related health problems. This fact may cause troubles when they wish to seek complete information regarding such problems. This study detects the associations among health problems by extending the meaning of health terms with methods based on the latent Dirichlet allocation (LDA) probability topic model, the Medical Subject Headings (MeSH) thesaurus structure and the Wikipedia concept mapping. The terms represented health problems are selected from and extended by the consumer-level medical text. The vocabulary is different between the consumer-level and the professional-level medical text. Thus, the findings can be easily understood by the general public and be suitable to consumer-oriented applications. The methods were evaluated in two ways: (1) correlation analysis with expert rating to show the overall performance and (2) P@N to reflect the ability of detecting strong associations. The LDA topic-model-based method outperforms the other two types. The judgment incongruence between the best method and the expert ratings has been examined, and the evidence shows that the automatic method sometimes detects real associations beyond those identified by human experts. [ABSTRACT FROM AUTHOR]

General news recommendations are important but have received limited attention because of the difficulties of measuring public interest. In public search engines, the objects of search terms reflect the issues that interest or concern search engine users. Because of the popularity of search engines, search indexes have become a new measure for describing public interest trends. With the help of a public search index provided by search engines, we construct a news topic search feature and a news object search feature. These features measure the public attention on key elements of the news. In the experiment, we compare various feature models with machine learning algorithms with respect to financial news recommendations. The results demonstrate that the topic search features perform best compared with other feature models. This research contributes to both the feature generation and news recommendation domains. [ABSTRACT FROM AUTHOR]

In recent years many automated topic coherence formulas (using the top-m words of a topic inferred by latent Dirichlet allocation) based on word similarities have been proposed and evaluated against human ratings. We treat a wordy topic as an object and quantitatively describe it via normalized mean values of pair-wise word similarities. Two types of word similarities, thesaurus and local corpus-based, are used as the descriptive features of a topic. We perform topic classification using represented topics as input and bi-level human ratings about topic coherence as class labels. Classification results (precision, recall and accuracy) based on two datasets and three supervised classification algorithms suggest that the novel topic representation is consistent with human ratings. Corpus-based word similarities are positively correlated with human ratings whereas thesaurus-based similarities have negative relations. The proposed representation of topics opens a window for us to investigate the utilization of topics with different perspectives. [ABSTRACT FROM AUTHOR]

As an important issue in sentiment analysis, sentence-level polarity classification plays a critical role in many opinion-mining applications such as opinion question answering, opinion retrieval and opinion summarization. Employing a supervised learning paradigm to train a classifier from sentences often faces the data sparseness problem owing to the short-length limit introduced to texts. In this article, regarding this problem, we exploit two different feature sets learned from external data sets as additional features to enrich data representation: one is a latent topic feature set obtained using a topic model, and the other is a related word feature set derived using word embeddings. Furthermore, we propose an ensemble approach by using these additional features to guide the design of different members of the ensemble. Experimental results on the public movie review dataset demonstrate that the enriched representations are effective for improving the performance of polarity classification, and the proposed ensemble approach can further improve the overall performance. [ABSTRACT FROM AUTHOR]

Probabilistic topic models are statistical methods whose aim is to discover the latent structure in a large collection of documents. The intuition behind topic models is that, by generating documents by latent topics, the word distribution for each topic can be modelled and the prior distribution over the topic learned. In this paper we propose to apply this concept by modelling the topics of sentences for the aspect detection problem in review documents in order to improve sentiment analysis systems. Aspect detection in sentiment analysis helps customers effectively navigate into detailed information about their features of interest. The proposed approach assumes that the aspects of words in a sentence form a Markov chain. The novelty of the model is the extraction of multiword aspects from text data while relaxing the bag-of-words assumption. Experimental results show that the model is indeed able to perform the task significantly better when compared with standard topic models. [ABSTRACT FROM PUBLISHER]

The Doubly Correlated Topic Model is a generative probabilistic topic model for automatically identifying topics from the corpus of the text documents. It is a mixed membership model, based on the fact that a document exhibits a number of topics. We used word co-occurrence statistical information for identifying an initial set of topics as posterior information for the model. Posterior inference methods utilized by the existing models are intractable and therefore provide an approximate solution. Consideration of co-occurred words as initial topics provides a tighter bound on the topic coherence. The proposed model is motivated by the Latent Dirichlet Allocation Model. The Doubly Correlated Topic Model differs from the Latent Dirichlet Allocation Model in its posterior inference; it uses the highest ranked co-occurred words as initial topics rather than obtaining from Dirichlet priors. The results of the proposed model suggest some improved performance on entropy and topical coherence over different datasets. [ABSTRACT FROM PUBLISHER]