Positive, Negative, or Mixed? Mining Blogs for Opinions

Third International AAAI Conference on Weblogs and …, 2009

This paper systematically exploited various lexical features for opinion analysis on blog data using a statistical learning framework. Our experimental results using the TREC Blog track data show that all the features we explored effectively represent opinion expressions, and different classification strategies have a significant impact on opinion classification performance. We also present results when combining opinion analysis with the retrieval component for the task of retrieving relevant and opinionated blogs. Compared with the best results in the TREC evaluation, our system achieves reasonable performance, but does not rely on much human knowledge or deep level linguistic analysis.

… : The 2010 Annual Conference of the North …, 2010

In this paper we examine different linguistic features for sentimental polarity classification, and perform a comparative study on this task between blog and review data. We found that results on blog are much worse than reviews and investigated two methods to improve the performance on blogs. First we explored information retrieval based topic analysis to extract relevant sentences to the given topics for polarity classification. Second, we adopted an adaptive method where we train classifiers from review data and incorporate their hypothesis as features. Both methods yielded performance gain for polarity classification on blog data.

2010

On lots of commercial blogs, bloggers' negative comments about enterprisers' images or harmful evaluations of products spread quickly in the cyber space. Moreover, an exposure of an inside story in the blogosphere may influence a company's reputation. Therefore, to identify bloggers' sentiment effectively is extremely important for enterprisers. These negative comments often bring great damage to enterprises. Recently, researchers proposed lots of machine learning techniques to efficiently detect customers' negative emotions for helping companies to carefully response customers' comments. However, they don't consider the class imbalance problem which lots of bloggers' comments are positive and far fewer comments are negative. A classifier induced from an imbalanced data set has high classification accuracy for the majority class, but an unacceptable error rate for the minority class. Therefore, this study proposed two new methods, MCBS and VS to provide a possible solution. Finally, a real case of bloggers' comments regarding MP3 products will be employed to illustrate the effectiveness of our proposed methods.

DyNaK 2010 Dynamic …, 2010

Abstract. The classification of a text according to its sentiment is a task of raising relevance in many applications, including applications related to monitoring and tracking of the blogosphere. The blogosphere provides a rich source of information about products, ...

2009 International Conference on Natural Language Processing and Knowledge Engineering, 2009

In this paper we explore the task of mood classification for blog postings. We propose a novel approach that uses the hierarchy of possible moods to achieve better results than a standard machine learning approach. We also show that using sentiment orientation features improves the performance of classification. We used the Livejournal blog corpus as a dataset to train and evaluate our method.

2013

Abstract—In this paper we explore the task of mood classification for blog postings. We propose a novel approach that uses the hierarchy of possible moods to achieve better results than a standard machine learning approach. We also show that using sentiment orientation features improves the performance of classification. We used the Livejournal blog corpus as a dataset to train and evaluate our method. Keywords:

2006

Intent mining is a special kind of document analysis whose goal is to assess the attitude of the document author with respect to a given subject. Opinion mining is a kind of intent mining where the attitude is a positive or negative opinion. Most systems tackle the problem with a two step approach, an information retrieval followed by a postprocess or filter phase to identify opinionated blogs. We explored a single stage approach to opinion mining, retrieving opinionated documents ranked with a special ranking function which exploits an index enriched with opinion tags. A set of subjective words are used as tags for identifying opinionated sentences. Subjective words are marked as "opinionated" and are used in the retrieval phase to boost the rank of documents containing them. In indexing the collection, we recovered the relevant content from the blog permalink pages, exploiting HTML metadata about the generator and heuristics to remove irrelevant parts from the body. The index also contains information about the occurrence of opinionated words, extracted from an analysis of WordNet glosses. The experiments compared the precision of normal queries with respect to queries which included as constraint the proximity to an opinionated word. The results show a significant improvement in precision for both topic relevance and opinion relevance.

2007

University of Arkansas at Little Rock's Blog Track team participated in only the core task of the blog track this year. The data acquired was identical to that of previous year except some new .retrieval tasks were introduced. The core task was to identify blogs that are opinionated about a certain subject. Fifty new topics were provided by National Institute of Standards and Technology (NIST) this year. Apart from the core task, two subtasks were also introduced. Polarity subtask was to detect polarity of the opinionated blog about a given topic. Feed distillation subtask was based on finding feeds rather than individual permalinks. Last year, we participated in the core task [1] and this year we planned to continue on our previous work. Although an attempt was made last year to use Active Learning with Support Vector Machine (SVM) to detect opinionated blog, identifying the opinion expressed about a given topic was unsuccessful. The difference this time around is in the use of search engines to conduct the topic search, categorizations of queries for further training, and a Natural Language based "onepass-processing" approach.

Handbook of Research on Soft Computing and Nature-Inspired Algorithms, 2017

Online social networking platforms, such as Weblogs, micro blogs, and social networks are intensively being utilized daily to express individual's thinking. This permits scientists to collect huge amounts of data and extract significant knowledge regarding the sentiments of a large number of people at a scale that was essentially impractical a couple of years back. Therefore, these days, sentiment analysis has the potential to learn sentiments towards persons, object and occasions. Twitter has increasingly become a significant social networking platform where people post messages of up to 140 characters known as ‘Tweets'. Tweets have become the preferred medium for the marketing sector as users can instantly indicate customer success or indicate public relations disaster far more quickly than a web page or traditional media does. In this paper, we have analyzed twitter data and have predicted positive and negative tweets with high accuracy rate using support vector machine (...

Now a day's sentiment analysis performs a very vital role in text mining. In essence web mining is a very broad area in a data mining field for extracts the sentiment of the text. To identify the sentiment of the textual data is a very challenging task. The present work focuses on sentence level negation identification and calculation from the News articles and Blogs. Two step approaches generally used for analysis namely preprocessing and post processing. Preprocessing consists of the tasks like stop word removing, punctuation mark removal, number removal, white space removal etc. Post processing comprises identification of sentiments from the text and calculation of score. The work analyses the performance of support vector machine, Naïve Bayes for the dataset collected online.