Leveraging Web 2.0 data for scalable semi-supervised learning of domain-specific sentiment lexicons

Since manually constructing domain-specific sentiment lexicons is extremely time consuming and it may not even be feasible for domains where linguistic expertise is not available. Research on the automatic construction of domain-specific sentiment lexicons has become a hot topic in recent years. The main contribution of this paper is the illustration of a novel semi-supervised learning method which exploits both term-to-term and document-to-term relations hidden in a corpus for the construction of domain specific sentiment lexicons. More specifically, the proposed two-pass pseudo labeling method combines shallow linguistic parsing and corpusbase statistical learning to make domain-specific sentiment extraction scalable with respect to the sheer volume of opinionated documents archived on the Internet these days. Another novelty of the proposed method is that it can utilize the readily available user-contributed labels of opinionated documents (e.g., the user ratings of product reviews) to bootstrap the performance of sentiment lexicon construction. Our experiments show that the proposed method can generate high quality domain-specific sentiment lexicons as directly assessed by human experts. Moreover, the system generated domain-specific sentiment lexicons can improve polarity prediction tasks at the document level by 2:18% when compared to other well-known baseline methods. Our research opens the door to the development of practical and scalable methods for domain-specific sentiment analysis.

Learning domain-specific sentiment lexicons for predicting product sales

Generic sentiment lexicons have been widely used for sentiment analysis these days. However, manually constructing sentiment lexicons is very time-consuming and it may not be feasible for certain application domains where annotation expertise is not available. One contribution of this paper is the development of a statistical learning based computational method for the automatic construction of domain-specific sentiment lexicons to enhance cross-domain sentiment analysis. Our initial experiments show that the proposed methodology can automatically generate domain-specific sentiment lexicons which contribute to improve the effectiveness of opinion retrieval at the document level. Another contribution of our work is that we show the feasibility of applying the sentiment metric derived based on the automatically constructed sentiment lexicons to predict product sales of certain product categories. Our research contributes to the development of more effective sentiment analysis system to extract business intelligence from numerous opinionated expressions posted to the Web

Online environmental citizenship : blogs, green marketing and consumer sentiment in the 21st Century

Over the last three decades, the rise in consumer generated content has enabled more environmentally conscious points of view to effect mainstream opinion (Kalafatis, Pollard, East & Tsogas, 1999; Barber, Taylor & Strick, 2009). Consequently, more people are buying into environmentalist ideology and organizing themselves to influence social change. Focus has shifted from attracting public awareness to concern for green ideas, discourse, and environmental citizenship, the latter becoming the guideline by which debates on such topics are regulated (Follows & Jobber, 2000; Dobson, 2003).

Integration of sentiment analysis into customer relational model : the importance of feature ontology and synonym

Online business or Electronic Commerce (EC) is getting popular among customers today, as a result large number of product reviews have been posted online by the customers. This information is very valuable not only for prospective customers to make decision on buying product but also for companies to gather information of customers’ satisfaction about their products. Opinion mining is used to capture customer reviews and separated this review into subjective expressions (sentiment word) and objective expressions (no sentiment word). This paper proposes a novel, multi-dimensional model for opinion mining, which integrates customers’ characteristics and their opinion about any products. The model captures subjective expression from product reviews and transfers to fact table before representing in multi-dimensions named as customers, products, time and location. Data warehouse techniques such as OLAP and Data Cubes were used to analyze opinionated sentences. A comprehensive way to calculate customers’ orientation on products’ features and attributes are presented in this paper.

Predicting stock market returns and volatility with investor sentiment : evidence from eight developed countries

We test the predictive ability of investor sentiment on the return and volatility at the aggregate market level in the U.S., four largest European countries and three Asia-Pacific countries. We find that in the U.S., France and Italy periods of high consumer confidence levels are followed by low market returns. In Japan both the level and the change in consumer confidence boost the market return in the next month. Further, shifts in sentiment significantly move conditional volatility in most of the countries, and in Italy such impacts lead to an increase in returns by 4.7% in the next month.

Investor sentiment as conditioning information in asset pricing

This paper assesses whether incorporating investor sentiment as conditioning information in asset-pricing models helps capture the impacts of the size, value, liquidity and momentum effects on risk-adjusted returns of individual stocks. We use survey sentiment measures and a composite index as proxies for investor sentiment. In our conditional framework, the size effect becomes less important in the conditional CAPM and is no longer significant in all the other models examined. Furthermore, the conditional models often capture the value, liquidity and momentum effects.

A non-syntactic approach for text sentiment classification with stopwords

The present approach uses stopwords and the gaps that oc- cur between successive stopwords –formed by contentwords– as features for sentiment classification.

The impact of investor sentiment on stock returns in emerging markets. The case of Central European markets

This is an accepted manuscript of an article published by Taylor & Francis in Eastern European Economics on July 2015, available online: http://dx.doi.org/10.1080/00128775.2015.1079139

"Out of sync": The breakdown of economic sentiment cycles in the E.U.

We present empirical evidence about the properties of economic sentiment cycle synchronization for Germany, France and the UK and compare them with the `crisis' countries Italy, Spain, Portugal and Greece. Instead of using output data we prefer to focus on the economic sentiment indicator (ESI), a forward-looking, survey-based variable consistently available from 1985. The cyclical nature of the ESI allows us to analyze the presence or not of synchronicity among country pairs before and after the onset of the financial crisis. Our results show that ESI movements were mostly synchronous before 2008 but they exhibit a breakdown after 2008, with this feature being more prominent in Greece. We also find that, after the political manoeuvring of the past two years, a cycle re-integration or re-synchronization is on the way. An analysis of the evolution of the synchronicity measures indicates that they can potentially be used to identify sudden phase breaks in ESI co-movement and they can offer a signal as to when the EU economies are getting “in” or “out of sync”.

‘Who Ate All the Pride?’:Patriotic Sentiment and English National Football Support.

The growing popularity of English national insignia in international football tournaments has been widely interpreted as evidence of the emergence of a renewed English national consciousness. However, little empirical research has considered how people in England actually understand football support in relation to national identity. Interview data collected around the time of the Euro 2000 and the 2002 World Cup tournaments fail to substantiate the presumption that support for the England football team maps onto claims to patriotic sentiment in any straightforward way. People with far-right political affiliations did generally use national football support to symbolise a general pride in English national identity. However, other people either claimed not to support the England national team precisely because of its associations with nationalism, or else bracketed the domain of football support from more general connotations of English patriotism.

Using Brain Data for Sentiment Analysis

We present the results of exploratory experiments using lexical valence extracted from brain using electroencephalography (EEG) for sentiment analysis. We selected 78 English words (36 for training and 42 for testing), presented as stimuli to 3 English native speakers. EEG signals were recorded from the subjects while they performed a mental imaging task for each word stimulus. Wavelet decomposition was employed to extract EEG features from the time-frequency domain. The extracted features were used as inputs to a sparse multinomial logistic regression (SMLR) classifier for valence classification, after univariate ANOVA feature selection. After mapping EEG signals to sentiment valences, we exploited the lexical polarity extracted from brain data for the prediction of the valence of 12 sentences taken from the SemEval-2007 shared task, and compared it against existing lexical resources.

Improved lexicon-based sentiment analysis for social media analytics

Social media channels, such as Facebook or Twitter, allow for people to express their views and opinions about any public topics. Public sentiment related to future events, such as demonstrations or parades, indicate public attitude and therefore may be applied while trying to estimate the level of disruption and disorder during such events. Consequently, sentiment analysis of social media content may be of interest for different organisations, especially in security and law enforcement sectors. This paper presents a new lexicon-based sentiment analysis algorithm that has been designed with the main focus on real time Twitter content analysis. The algorithm consists of two key components, namely sentiment normalisation and evidence-based combination function, which have been used in order to estimate the intensity of the sentiment rather than positive/negative label and to support the mixed sentiment classification process. Finally, we illustrate a case study examining the relation between negative sentiment of twitter posts related to English Defence League and the level of disorder during the organisation’s related events.