We propose a semi-supervised framework for generating a domain-specific sentiment lexicon and inferring sentiments at the segment level.

Our framework can greatly reduce the human effort for building a domain-specific sentiment lexicon with high quality.

Specifically, in our evaluation, working with just 20 manually labeled reviews, it generates a domain-specific sentiment lexicon that yields weighted average F-Measure gains of 3%.

A domain-specific sentiment lexicon con-

Given a set of reviews, the tasks of sentiment analysis in ReNew are (l) splitting each review into segments, (2) associating each segment with a sentiment label (positive, neutral, negative), and (3) automatically generating a domain-specific sentiment lexicon .

a general sentiment lexicon and a small labeled training dataset.

High-quality sentiment lexicons can improve the performance of sentiment analysis models over general-purpose lexicons (Choi and Cardie, 2009).

(2) ReNew leverages the relationships between consecutive segments to infer their sentiments and automatically generates a domain-specific sentiment lexicon in a semi-su-pervised fashion.

(3) To capture the contextual sentiment of words, ReNew uses dependency relation pairs as the basic elements in the generated sentiment lexicon .

(2013) build the top-performed system in the Twitter sentiment classification track of SemEval 2013 (Nakov et al., 2013), using diverse sentiment lexicons and a variety of handcrafted features.

The quality of SSWE is also directly evaluated by measuring the word similarity in the embedding space for sentiment lexicons .

We also directly evaluate the effectiveness of the SSWE by measuring the word similarity in the embedding space for sentiment lexicons .

(5) NRC: NRC builds the top-performed system in SemEval 2013 Twitter sentiment classification track which incorporates diverse sentiment lexicons and many manually designed features.

We achieve 84.98% by using only SSWEu as features without borrowing any sentiment lexicons or handcrafted rules.

Sentiment analysis in a multilingual world remains a challenging problem, because developing language-specific sentiment lexicons is an extremely resource-intensive process.

In this paper, we address this lexicon gap by building high-quality sentiment lexicons for 136 major languages.

By appropriately propagating from seed words, we construct sentiment lexicons for each component language of our graph.

Sentiment propagation starts from English sentiment lexicons .

Although several well-regarded sentiment lexicons are available in English (Esuli and Sebastiani, 2006; Liu, 2010), the same is not true for most of the world’s languages.

Indeed, our literature search identified only 12 publicly available sentiment lexicons for only 5 non-English languages (Chinese mandarin, German, Arabic, Japanese and

In this paper, we strive to produce a comprehensive set of sentiment lexicons for the worlds’ major languages.

In this section we will describe how we leverage off a variety of NLP resources to construct the semantic connection graph we will use to propagate sentiment lexicons .

(2010) focus on generating topic specific sentiment lexicons .

Starting with a domain-independent, high-precision sentiment lexicon and a large pool of unlabeled data, we bootstrap Twitter-specific sentiment lexicons , using a small amount of labeled data to guide the process.

General, domain-independent sentiment lexicons have low coverage.

Most of the previous work on sentiment lexicon construction relies on existing natural language

Although bootstrapping has been used for leam-ing sentiment lexicons in other domains (Turney and Littman, 2002; Banea et al., 2008), it has not yet been applied to learning sentiment lexicons for microblogs.

To create a Twitter-specific sentiment lexicon for a given language, we start with a general-purpose, high-precision sentiment lexicon2 and bootstrap from the unlabeled data (BOOT) using the labeled development data (DEV) to guide the process.

Dictionary-based methods rely on existing lexical resources to bootstrap sentiment lexicons .

(2009) use a thesaurus to aid in the construction of a sentiment lexicon for English.

Corpus-based methods extract subjectivity and sentiment lexicons from large amounts of unlabeled data using different similarity metrics to measure the relatedness between words.

We develop induction algorithms based on three distinct types of algorithmic framework that have been shown successful for the analogous task of sentiment lexicon induction: HITS & PageRank (§2.1), Label/Graph Propagation (§2.2), and Constraint Optimization via Integer Linear Programming (§2.3).

3.1 Comparison against Conventional Sentiment Lexicon

Note that we consider the connotation lexicon to be inclusive of a sentiment lexicon for two practical reasons: first, it is highly unlikely that any word with non-neutral sentiment (i.e., positive or negative) would carry connotation of the opposite, i.e., conflicting10 polarity.

Therefore, sentiment lexicons can serve as a surrogate to measure a subset of connotation words induced by the algorithms, as shown in Table 3 with respect to General Inquirer (Stone and Hunt (1963)) and MPQA (Wilson et al.

In this section, we present comprehensive intrinsic §5.l and extrinsic §5 .2 evaluations comparing three representative lexicons from §2 & §42 C-LP, OVERLAY, PRED-ARG (CP), and two popular sentiment lexicons : SentiWordNet (Baccianella et al., 2010) and GI+MPQA.14 Note that C-LP is the largest among all connotation lexicons, including ~70,000 polar words.15

The main contribution of this paper is a broad-coverage connotation lexicon that determines the connotative polarity of even those words with ever so subtle connotation beneath their surface meaning, such as “Literature”, “Mediterranean”, and “wine Although there has been a number of previous work that constructed sentiment lexicons (e.g., Esuli and Sebastiani (2006), Wilson et al.

Although such an assumption played a key role in previous work for the analogous task of learning sentiment lexicon (Velikovich et al., 2010), we expect that the same assumption would be less reliable in drawing subtle connotative sentiments of words.

We cast the connotation lexicon induction task as a collective inference problem, and consider approaches based on three distinct types of algorithmic framework that have been shown successful for conventional sentiment lexicon induction:

13 Note that doing so will prevent us from evaluating against the same sentiment lexicon used as a seed set.

Some recent work explored the use of constraint optimization framework for inducing domain-dependent sentiment lexicon (Choi and Cardie (2009), Lu et al.

Sentiment lexicon construction and topic lexicon extraction are two fundamental subtasks for opinion mining (Qiu et al., 2009).

A sentiment lexicon is a list of sentiment expressions, which are used to indicate sentiment polarity (e.g., positive or negative).

The sentiment lexicon is domain dependent as users may use different sentiment words to express their opinion in different domains (e. g., different products).

Our goal now is to bias these models with constraints incorporating (a) labels of features (coming from a domain-independent sentiment lexicon ), and (b) labels of documents for the purposes of domain-specific adaptation.

Our interest in the first set of experiments is to explore the benefits of incorporating a sentiment lexicon over unsupervised approaches.

These methods do not make use of the sentiment lexicon .

Size of Sentiment Lexicon We also investigate the effects of the size of the sentiment lexicon on the performance of our model.

We used a sentiment lexicon generated by the IBM India Research Labs that was developed for other text mining applications (Ramakrishnan et al., 2003).

Treated as a set of labeled features, the sentiment lexicon is incorporated as one set of constraints that enforce domain-independent prior knowledge.

So far our models have made no demands on human effort, other than unsupervised collection of the term-document matrix and a onetime effort in compiling a domain-independent sentiment lexicon .

We present comprehensive evaluation to demonstrate the quality and utility of the resulting lexicon in comparison to existing connotation and sentiment lexicons .

For comparison, we also test the connotation lexicon from (Feng et al., 2013) and the combined sentiment lexicon GENINQ+MPQA.

Sentiment lexicons such as SentiWordNet (Baccianella et al.

We also extended the seed set with the sentiment lexicon words and denote these runs with E- for ‘Extended’.

ConnotationWordNet is expected to be the superset of a sentiment lexicon , as it is highly likely for any word with positive/negative sentiment to carry connotation of the same polarity.

Thus, we use two conventional sentiment lexicons , General Inquirer (GENINQ) (Stone et al., 1966) and MPQA (Wilson et al., 2005b), as surrogates to measure the performance of our inference algorithm.

The sentiment lexicons we use as gold standard are small, compared to the size (i.e., number of words) our graphs contain.

Several previous approaches explored the use of graph propagation for sentiment lexicon induction (Velikovich et al., 2010) and connotation lexicon

There have been a number of previous studies that aim to construct a word-level sentiment lexicon (Wiebe et al., 2005; Qiu et al., 2009) and a sense-level sentiment lexicon (Esuli and Sebas-tiani, 2006).

Although we focus on learning connotative polarity of words and senses in this paper, the same approach would be applicable to constructing a sentiment lexicon as well.

Sentiment lexicon features, indicating how many positive or negative words are included in the tweet according to a predefined lexicon.

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)

Adding sentiment lexicon features improves the accuracy to 63.8%.

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)

We conjecture that sentiment prediction for AUTO category is largely driven by one-shot phrases and statements where it is hard to improve upon the bag-of-words and sentiment lexicon features.

We enrich the traditional bag-of-word representation with features from a sentiment lexicon and features quantifying the negation present in the comment.

- lexicon: a sentiment lexicon is a collection of words associated with a positive or negative sentiment.

We use two manually constructed sentiment lexicons that are freely available: the MPQA Lexicon (Wilson et al., 2005) and the lexicon of Hu and Liu (2004).

We demonstrate our approach on a new dataset just released by Yelp, and release a new sentiment lexicon with 1329 adjectives for the restaurant domain.

Using topic models to discover subtypes of businesses, a domain-specific sentiment lexicon , and a number of new techniques for increasing precision in sentiment aspect extraction yields attributes that give a rich representation of the restaurant domain.

We have made this l329-term sentiment lexicon for the restaurant domain available as useful resource to the community.

First we develop a domain specific sentiment lexicon .

2.2.1 Sentiment Lexicon

To identify noun-phrases which are targeted by predicates in our sentiment lexicon , we develop handcrafted extraction patterns defined over syntactic dependency parses (Blair—Goldensohn et al., 2008; Somasundaran and Wiebe, 2009) generated by the Stanford parser (Klein and Manning, 2003).

Such work generally exploits textual features for fact-based analysis tasks or lexical indicators from a sentiment lexicon .

Accordingly, much research has focused on recognizing terms’ semantic orientations and strength, and compiling sentiment lexicons (Hatzivassiloglou and Mckeown, 1997; Turney and Littman, 2003; Kamps et al., 2004; Whitelaw et al., 2005; Esuli and Sebastiani, 2006).

The goal of this paper is not to create or choose an appropriate sentiment lexicon , but rather it is to discover useful term features other than the sentiment properties.

For this reason, one sentiment lexicon , namely SentiWordNet, is utilized throughout the whole experiment.

SentiWordNet is an automatically generated sentiment lexicon using a semi-supervised method (Esuli and Sebastiani, 2006).

In the following paper, we describe our attempts to use modern sentiment lexicons and dialogue structure to algorithmically track and model—with no domain-specific customization—the emotion dynamics between characters in Shakespeare’s

These methods are driven by sentiment lexicons , fixed lists associating words with “valences” (signed integers representing positive and negative feelings) (Kim and Hovy, 2004).

To extract these relationships, we mined for character-to-character sentiment by summing the valence values (provided by the AFINN sentiment lexicon (Nielsen, 2011)) over each instance of continuous speech and then assumed that sentiment was directed towards the character that spoke immediately before the current speaker.

The previously proposed J ST model uses the sentiment prior information in the Gibbs sampling inference step that a sentiment label will only be sampled if the current word token has no prior sentiment as defined in a sentiment lexicon .

For each word 21) E {1, ..., V}, if w is found in the sentiment lexicon , for each I E {1, ..., S}, the element Alw is updated as follows

where the function 8 returns the prior sentiment label of w in a sentiment lexicon , i.e.

The MPQA subjectivity lexicon is used as a sentiment lexicon in our experiments.

5.3 Sentiment Lexicon Induction

(2007) propose two methods for translating sentiment lexicons .

The first method simply uses bilingual dictionaries to translate an English sentiment lexicon .

The induction of a sentiment lexicon is the subject of early work by (Hatzivassiloglou and McKeown, 1997).

To learn the Max-Ent parameters 2 of MESAS, we used the sentiment lexicon 4 of (Hu and Liu, 2004) to automatically generate training data (no manual labeling).

Of those 1000 terms if they appeared in the sentiment lexicon , they were treated as sentiment terms, else aspect terms.

Clearly, labeling words not in the sentiment lexicon as aspect terms may not always be correct.

The features used by these models include subjectivity and sentiment lexicons , counts of unigrams and bigrams, distributional similarity, discourse relationships, and so on.

Of the 654 words included in this lexicon, 433 were unique to this lexicon (i.e., recorded in neither Riloff and Wiebe’s (2003) subjectivity lexicon nor Liu et al.’s (2005) sentiment lexicon ) and represented many one-sided or controversial terms, e.g., abortion, same-sex, execute.

To this end, the features based on subjectivity and sentiment lexicons turn out to be helpful, and incorporating more features for stance detection is an important direction for future work.

Prior knowledge of the sentiment of words, such as sentiment lexicons , has been incorporated into cross-domain sentiment classification.

(2011) propose a joint sentiment-topic model that imposes a sentiment-prior depending on the occurrence of a word in a sentiment lexicon .

A sentiment lexicon is used to create features for a document.

Sentiments are extracted using a sentiment lexicon and pattern matched from dependency trees of sentences.

Note that since our method relies on only sentiment lexicon , extractable aspects are unlimited.

1Since we aim to summarize Japanese reviews, we utilize Japanese sentiment lexicon (Asano et al., 2008).

where O is the set of terms in the sentiment lexicon , P(sub|w) indicates the probability of term 212 being subjective, and n(w, D) is the number of times term 21) occurs in document D. The opinion scoring can weigh lexicon terms differently, using P(sub|w); it normalizes scores to cancel out the effect of varying document sizes.

where n(O, D) is the number of matches of the term of sentiment lexicon O in document D.

Since it is unrealistic to construct sentiment lexicons , or manually annotate text for learning, for every imaginable domain or topic, automatic methods have been developed.