Semi-supervised vs. Cross-domain Graphs for Sentiment Analysis
نویسندگان
چکیده
The lack of labeled data always poses challenges for tasks where machine learning is involved. Semi-supervised and cross-domain approaches represent the most common ways to overcome this difficulty. Graph-based algorithms have been widely studied during the last decade and have proved to be very effective at solving the data limitation problem. This paper explores one of the most popular stateof-the-art graph-based algorithms label propagation, together with its modifications previously applied to sentiment classification. We study the impact of modified graph structure and parameter variations and compare the performance of graph-based algorithms in cross-domain and semi-supervised settings. The results provide a strategy for selecting the most favourable algorithm and learning paradigm on the basis of the available labeled and unlabeled data.
منابع مشابه
Graph-based approaches for semi-supervised and cross-domain sentiment analysis
The rapid development of Internet technologies has resulted in a sharp increase in the number of Internet users who create content online. Usergenerated content often represents people’s opinions, thoughts, speculations and sentiments and is a valuable source of information for companies, organisations and individual users. This has led to the emergence of the field of sentiment analysis, which...
متن کاملA Supervised Method for Constructing Sentiment Lexicon in Persian Language
Due to the increasing growth of digital content on the internet and social media, sentiment analysis problem is one of the emerging fields. This problem deals with information extraction and knowledge discovery from textual data using natural language processing has attracted the attention of many researchers. Construction of sentiment lexicon as a valuable language resource is a one of the imp...
متن کاملCustomizing Sentiment Classifiers to New Domains: a Case Study
Sentiment classification is a very domainspecific problem; classifiers trained in one domain do not perform well in others. Unfortunately, many domains are lacking in large amounts of labeled data for fully-supervised learning approaches. At the same time, sentiment classifiers need to be customizable to new domains in order to be useful in practice. We attempt to address these difficulties and...
متن کاملUser-guided Cross-domain Sentiment Classification
Sentiment analysis has been studied for decades, and it is widely used in many real applications such as media monitoring. In sentiment analysis, when addressing the problem of limited labeled data from the target domain, transfer learning, or domain adaptation, has been successfully applied, which borrows information from a relevant source domain with abundant labeled data to improve the predi...
متن کاملReNew: A Semi-Supervised Framework for Generating Domain-Specific Lexicons and Sentiment Analysis
The sentiment captured in opinionated text provides interesting and valuable information for social media services. However, due to the complexity and diversity of linguistic representations, it is challenging to build a framework that accurately extracts such sentiment. We propose a semi-supervised framework for generating a domain-specific sentiment lexicon and inferring sentiments at the seg...
متن کامل