Word to sentence level emotion tagging for Bengali blogs

Das, Dipankar; Bandyopadhyay, Sivaji

doi:10.3115/1667583.1667630

Cited by 45 publications

(25 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, it also shows that feature selection has a great influence on the final results. According to the research result, the combination of multiple features in comparison with a single feature generally shows a reasonable enhancement of any classification system [6]. Consequently, through manually reviewing the short comments corpus and their language specific characteristics, the feature sets for opinion objects are defined, as shown in the table 1.…”

Section: Opinion Objects Identificationmentioning

confidence: 99%

Features-level Sentiment Analysis of Movie Reviews

Ouyang¹,

Liu²,

Zhang³

et al. 2015

Advanced Science and Technology Letters

View full text Add to dashboard Cite

Abstract. Most sentiment analysis of reviews focused on extracting overall evaluation for a single product which makes difficult for a customer to know all the features of product and make a decision. Thus, mining this data, identifying the user opinions about different features and classify them is an important task. This paper is devoted to identify opinion object from short comments of movies, and analyze sentiment of movies based on features-level. CRFs model based on word embedding feature is adopted by identifying opinion object. In addition, calculate rules based on syntax parsing are proposed to accomplish featureslevel sentiment analysis. Experimental results show the effectiveness of our approach.

show abstract

Section: Opinion Objects Identificationmentioning

confidence: 99%

Features-level Sentiment Analysis of Movie Reviews

Ouyang¹,

Liu²,

Zhang³

et al. 2015

Advanced Science and Technology Letters

View full text Add to dashboard Cite

show abstract

“…The non emotional sentences are considered as neutral. We have employed a sentential emotion tagging system [11] that consists of two prong approach, word level followed by sentence level. The Conditional Random Field (CRF) [19] based machine learning approach that incorporates several singleton features (e.g.…”

Section: Automatic Emotion Taggingmentioning

confidence: 99%

“…The basic six words "happy", "sad", "anger", "disgust", "fear" "surprise" are selected as the seed words corresponding to each emotion type. The positive and negative scores of each synset in which each of these seed words appear are retrieved from the English SentiWordNet [20] and the average of the scores is fixed as the Sense_Tag_Weight (STW) of that particular emotion tag (happy: 0.0125, sad: -0.1022, anger: -0.5, disgust: -0.075, fear: 0.0131, surprise: 0.0625, and neutral: 0.0) [11].…”

Section: Automatic Emotion Taggingmentioning

confidence: 99%

“…The identification of Ekman's [23] six basic emotions from the bloggers' comments is carried out at sentence and paragraph level granularities using the information of word level constituents along with sense based scoring mechanism [11]. An affect scoring technique has been employed to identify the emotions of a state or node in each of the Referential Informative Chains (RICs) of the bloggers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Tracking Emotions of Bloggers - A Case Study for Bengali

Das¹,

Bandyopadhyay²

2012

Polibits

Self Cite

View full text Add to dashboard Cite

Abstract-The present paper describes the identification and tracking of bloggers' emotions with respect to time from the structured Bengali blog documents. The assignment of Ekman's six basic emotions to the bloggers' comments is carried out at sentence and paragraph level granularities. The Referential Informative Chain (RIC) developed for each blogger consists of the nodes representing the emotional states of that blogger. Each node of a RIC contains the identification information of its associated blogger, timestamp, section and emotional sentences. The nodes are arranged in each RIC based on the ascending order of the associated timestamps. An affect scoring technique has been employed to capture the emotions from each of the nodes of a blogger's RIC. The incorporation of self emotions and influential emotions as extracted from other bloggers plays a significant role in detecting the emotions of a blogger's present state. The Extrinsic evaluation produces precision (P), recall (R) and F-Measure of 61.05%, 69.81% and 65.13% respectively for evaluating the total of 193 emotional states of 20 bloggers. The Intrinsic evaluation has been conducted using a manual rater with the help of a statistical agreement coefficient, Krippendorff's alpha α. Two types of alpha, namely nominal alpha and interval alpha produce the average scores of 0.67 and 0.72, respectively.

show abstract

“…Moreover, human feelings can be quickly identified through automatic emotion classification, as these emotions reflect an individual's feelings and experiences toward certain subject matters (Turney, 2002;Wilson et al, 2009). Emotion classification aims to predict the emotion categories (e.g., happy, angry, or worried) to which the given text belongs (Das and Bandyopadhyay, 2009;Quan and Ren, 2009). There are two aspects of emotions regarding a piece of text, namely, the writer's and the reader's emotion.…”

Section: Introductionmentioning

confidence: 99%

How Do I Look? Publicity Mining From Distributed Keyword Representation of Socially Infused News Articles

Hsieh

Chang

Chu

et al. 2016

Proceedings of the Fourth International Workshop on Natural Language Processing for Social Media

View full text Add to dashboard Cite

Previous work on opinion mining and sentiment analysis mainly concerns product, movie, or literature reviews; few applied this technique to analyze the publicity of person. We present a novel document modeling method that utilizes embeddings of emotion keywords to perform reader's emotion classification, and calculates a publicity score that serves as a quantifiable measure for the publicity of a person of interest. Experiments are conducted on two Chinese corpora that in total consists of over forty thousand users' emotional response after reading news articles. Results demonstrate that the proposed method can outperform state-ofthe-art reader-emotion classification methods, and provide a substantial ground for publicity score estimation for candidates of political elections. We believe it is a promising direction for mining the publicity of a person from online social and news media that can be useful for propaganda and other purposes.

show abstract

Word to sentence level emotion tagging for Bengali blogs

Cited by 45 publications

References 8 publications

Features-level Sentiment Analysis of Movie Reviews

Features-level Sentiment Analysis of Movie Reviews

Tracking Emotions of Bloggers - A Case Study for Bengali

How Do I Look? Publicity Mining From Distributed Keyword Representation of Socially Infused News Articles

Contact Info

Product

Resources

About