Target-dependent Twitter Sentiment Classification
Jiang, Long and Yu, Mo and Zhou, Ming and Liu, Xiaohua and Zhao, Tiejun

Article Structure

Abstract

Sentiment analysis on Twitter data has attracted much attention recently.

Introduction

Twitter, as a micro-blogging system, allows users to publish tweets of up to 140 characters in length to tell others what they are doing, what they are thinking, or what is happening around them.

Related Work

In recent years, sentiment analysis (SA) has become a hot topic in the NLP research community.

Approach Overview

The problem we address in this paper is target-dependent sentiment classification of tweets.

Target-dependent Sentiment Classification

Besides target-independent features, we also incorporate target-dependent features in both the subjec-

Graph-based Sentiment Optimization

As we mentioned in Section 1, since tweets are usually shorter and more ambiguous, it would be useful to take their contexts into consideration when classifying the sentiments.

Experiments

Because there is no annotated tweet corpus publicly available for evaluation of target-dependent Twitter sentiment classification, we have to create our own.

Conclusions and Future Work

Twitter sentiment analysis has attracted much attention recently.

Topics

sentiment classification

Appears in 32 sentences as: sentiment class (2) Sentiment Classification (2) sentiment classification (27) sentiment classifier (1)
In Target-dependent Twitter Sentiment Classification
  1. In this paper, we focus on target-dependent Twitter sentiment classification ; namely, given a query, we classify the sentiments of the tweets as positive, negative or neutral according to whether they conuun posfljve, negafive or nqual senfi-ments about that query.
    Page 1, “Abstract”
  2. However, because tweets are usually short and more ambiguous, sometimes it is not enough to consider only the current tweet for sentiment classification .
    Page 1, “Abstract”
  3. In this paper, we propose to improve target-dependent Twitter sentiment classification by 1) incorporating target-dependent features; and 2) taking related tweets into consideration.
    Page 1, “Abstract”
  4. According to the experimental results, our approach greatly im-proveslhe perfintnance of Hugebdependent sentiment classification .
    Page 1, “Abstract”
  5. 3r Sentiment Classification
    Page 1, “Introduction”
  6. The problem needing to be addressed can be formally named as Target-dependent Sentiment Classification of Tweets; namely, given a query, classifying the sentiments of the tweets as positive, negative or neutral according to whether they contain positive, negative or neutral sentiments about that query.
    Page 1, “Introduction”
  7. The state-of-the-art approaches for solving this problem, such as (Go et al., 20095; Barbosa and Feng, 2010), basically follow (Pang et al., 2002), who utilize machine learning based classifiers for the sentiment classification of texts.
    Page 1, “Introduction”
  8. Since (Pang et al., 2002) (or later research on sentiment classification
    Page 1, “Introduction”
  9. However, for target-dependent sentiment classification of tweets, it is not suitable to exactly adopt that approach.
    Page 2, “Introduction”
  10. However, with target-independent sentiment classification , both of the targets would get positive polarity.
    Page 2, “Introduction”
  11. In this paper, we propose to improve target-dependent sentiment classification of tweets by using both target-dependent and context-aware approaches.
    Page 2, “Introduction”

See all papers in Proc. ACL 2011 that mention sentiment classification.

See all papers in Proc. ACL that mention sentiment classification.

Back to top.

graph-based

Appears in 8 sentences as: +Graph-based (1) Graph-based (2) graph-based (5)
In Target-dependent Twitter Sentiment Classification
  1. Graph-based optimization as the third step to further boost the performance by taking the related tweets into consideration.
    Page 3, “Approach Overview”
  2. If we consider that the sentiment of a tweet only depends on its content and immediate neighbors, we can leverage a graph-based method for sentiment classification of tweets.
    Page 6, “Graph-based Sentiment Optimization”
  3. 6.4 Evaluation of Graph-based Optimization
    Page 8, “Experiments”
  4. For these tweets, our graph-based optimization approach will have no effect.
    Page 8, “Experiments”
  5. Target-dependent sentiment classifier +Graph-based optimization
    Page 8, “Experiments”
  6. The graph-based optimization improves the performance by over 2 points (p < 0.005), which clearly shows that the context information is very
    Page 8, “Experiments”
  7. In addition, different from previous work using only information on the current tweet for sentiment classification, we propose to take the related tweets of the current tweet into consideration by utilizing graph-based optimization.
    Page 9, “Conclusions and Future Work”
  8. According to the experimental results, the graph-based optimization significantly improves the performance.
    Page 9, “Conclusions and Future Work”

See all papers in Proc. ACL 2011 that mention graph-based.

See all papers in Proc. ACL that mention graph-based.

Back to top.

Sentiment lexicon

Appears in 8 sentences as: Sentiment lexicon (4) sentiment lexicon (4)
In Target-dependent Twitter Sentiment Classification
  1. Sentiment lexicon features, indicating how many positive or negative words are included in the tweet according to a predefined lexicon.
    Page 4, “Approach Overview”
  2. Features Accuracy (%) Content features 61.1 + Sentiment lexicon features 63.8 + Target-dependent features 68.2 Re-implementation of (Bar- 60.3 bosa and Feng, 2010)
    Page 7, “Experiments”
  3. Adding sentiment lexicon features improves the accuracy to 63.8%.
    Page 7, “Experiments”
  4. Features Accuracy (%) Content features 78.8 + Sentiment lexicon features 84.2 + Target-dependent features 85.6 Re-implementation of (Bar- 83.9 bosa and Feng, 2010)
    Page 8, “Experiments”
  5. Sentiment lexicon features are shown to be very helpful for improving the performance.
    Page 8, “Experiments”
  6. The results show that our system using both content features and sentiment lexicon features performs slightly better than (Barbosa and Feng, 2010).
    Page 8, “Experiments”
  7. Both the classifiers with all features and with the combination of content and sentiment lexicon features are significantly better than that with only the content features (p < 0.01).
    Page 8, “Experiments”
  8. However, the classifier with all features does not significantly outperform that using the combination of content and sentiment lexicon features.
    Page 8, “Experiments”

See all papers in Proc. ACL 2011 that mention Sentiment lexicon.

See all papers in Proc. ACL that mention Sentiment lexicon.

Back to top.

sentiment analysis

Appears in 7 sentences as: Sentiment analysis (1) sentiment analysis (6)
In Target-dependent Twitter Sentiment Classification
  1. Sentiment analysis on Twitter data has attracted much attention recently.
    Page 1, “Abstract”
  2. In fact, it is easy to find many such cases by looking at the output of Twitter Sentiment or other Twitter sentiment analysis web sites.
    Page 2, “Introduction”
  3. In addition, tweets are usually shorter and more ambiguous than other sentiment data commonly used for sentiment analysis , such as reviews and blogs.
    Page 2, “Introduction”
  4. In recent years, sentiment analysis (SA) has become a hot topic in the NLP research community.
    Page 2, “Related Work”
  5. As Twitter becomes more popular, sentiment analysis on Twitter data becomes more attractive.
    Page 3, “Related Work”
  6. Previous work (Barbosa and Feng, 2010; Davidiv et al., 2010) has discovered many effective features for sentiment analysis of tweets, such as emoticons, punctuation, prior subjectivity and polarity of a word.
    Page 4, “Approach Overview”
  7. Twitter sentiment analysis has attracted much attention recently.
    Page 9, “Conclusions and Future Work”

See all papers in Proc. ACL 2011 that mention sentiment analysis.

See all papers in Proc. ACL that mention sentiment analysis.

Back to top.

noun phrases

Appears in 4 sentences as: noun phrases (4)
In Target-dependent Twitter Sentiment Classification
  1. In this paper, we first regard all noun phrases , including the target, as extended targets for simplicity.
    Page 4, “Target-dependent Sentiment Classification”
  2. In addition to the noun phrases including the target, we further expand the extended target set with the following three methods:
    Page 4, “Target-dependent Sentiment Classification”
  3. It is common that people use definite or demonstrative noun phrases or pronouns referring to the target in a tweet and express sentiments directly on them.
    Page 4, “Target-dependent Sentiment Classification”
  4. Identifying the top K nouns and noun phrases which have the strongest association with the target.
    Page 5, “Target-dependent Sentiment Classification”

See all papers in Proc. ACL 2011 that mention noun phrases.

See all papers in Proc. ACL that mention noun phrases.

Back to top.

SVM

Appears in 4 sentences as: SVM (4)
In Target-dependent Twitter Sentiment Classification
  1. According to the experimental results, machine learning based classifiers outperform the unsupervised approach, where the best performance is achieved by the SVM classifier with unigram presences as features.
    Page 3, “Related Work”
  2. In contrast, (Barbosa and Feng, 2010) propose a two-step approach to classify the sentiments of tweets using SVM classifiers with abstract features.
    Page 3, “Related Work”
  3. In each of the first two steps, a binary SVM classifier is built to perform the classification.
    Page 3, “Approach Overview”
  4. In the experiments, we consider the positive and negative tweets annotated by humans as subjective tweets (i.e., positive instances in the SVM classifiers), which amount to 727 tweets.
    Page 7, “Experiments”

See all papers in Proc. ACL 2011 that mention SVM.

See all papers in Proc. ACL that mention SVM.

Back to top.

syntactic parsing

Appears in 4 sentences as: syntactic parse (1) syntactic parsing (3)
In Target-dependent Twitter Sentiment Classification
  1. For instance, in the second example, using syntactic parsing , we know that “Windows 7” is connected to “better” by a copula, while “Vista” is connected to “better” by a preposition.
    Page 2, “Introduction”
  2. In order to generate such features, much NLP work has to be done beforehand, such as tweet normalization, POS tagging, word stemming, and syntactic parsing .
    Page 4, “Approach Overview”
  3. For syntactic parsing we use a Maximum Spanning Tree dependency parser (McDonald et al., 2005).
    Page 4, “Approach Overview”
  4. In this paper, we rely on the syntactic parse tree to satisfy this need.
    Page 5, “Target-dependent Sentiment Classification”

See all papers in Proc. ACL 2011 that mention syntactic parsing.

See all papers in Proc. ACL that mention syntactic parsing.

Back to top.

machine learning

Appears in 3 sentences as: machine learning (3)
In Target-dependent Twitter Sentiment Classification
  1. The state-of-the-art approaches for solving this problem, such as (Go et al., 20095; Barbosa and Feng, 2010), basically follow (Pang et al., 2002), who utilize machine learning based classifiers for the sentiment classification of texts.
    Page 1, “Introduction”
  2. According to the experimental results, machine learning based classifiers outperform the unsupervised approach, where the best performance is achieved by the SVM classifier with unigram presences as features.
    Page 3, “Related Work”
  3. (Go et al., 2009; Parikh and Movassate, 2009; Barbosa and Feng, 2010; Davidiv et al., 2010) all follow the machine learning based approach for sentiment classification of tweets.
    Page 3, “Related Work”

See all papers in Proc. ACL 2011 that mention machine learning.

See all papers in Proc. ACL that mention machine learning.

Back to top.

PDF available from www.aclweb.org.
Send us an email