A comparative study of bayesian models for unsupervised sentiment detection

This paper presents a comparative study of three closely related Bayesian models for unsupervised document level sentiment classification, namely, the latent sentiment model (LSM), the joint sentiment-topic (JST) model, and the Reverse-JST model. Extensive experiments have been conducted on two corpora, the movie review dataset and the multi-domain sentiment dataset. It has been found that while all the three models achieve either better or comparable performance on these two corpora when compared to the existing unsupervised sentiment classification approaches, both JST and Reverse-JST are able to extract sentiment-oriented topics. In addition, Reverse-JST always performs worse than JST suggesting that the JST model is more appropriate for joint sentiment topic detection.

Learning sentiment classification model from labeled features

We propose a novel framework where an initial classifier is learned by incorporating prior information extracted from an existing sentiment lexicon. Preferences on expectations of sentiment labels of those lexicon words are expressed using generalized expectation criteria. Documents classified with high confidence are then used as pseudo-labeled examples for automatical domain-specific feature acquisition. The word-class distributions of such self-learned features are estimated from the pseudo-labeled examples and are used to train another classifier by constraining the model's predictions on unlabeled instances. Experiments on both the movie review data and the multi-domain sentiment dataset show that our approach attains comparable or better performance than exiting weakly-supervised sentiment classification methods despite using no labeled documents.

Automatically extracting polarity-bearing topics for cross-domain sentiment classification

Joint sentiment-topic (JST) model was previously proposed to detect sentiment and topic simultaneously from text. The only supervision required by JST model learning is domain-independent polarity word priors. In this paper, we modify the JST model by incorporating word polarity priors through modifying the topic-word Dirichlet priors. We study the polarity-bearing topics extracted by JST and show that by augmenting the original feature space with polarity-bearing topics, the in-domain supervised classifiers learned from augmented feature representation achieve the state-of-the-art performance of 95% on the movie review data and an average of 90% on the multi-domain sentiment dataset. Furthermore, using feature augmentation and selection according to the information gain criteria for cross-domain sentiment classification, our proposed approach performs either better or comparably compared to previous approaches. Nevertheless, our approach is much simpler and does not require difficult parameter tuning.

Semantic sentiment analysis of Twitter

Sentiment analysis over Twitter offer organisations a fast and effective way to monitor the publics' feelings towards their brand, business, directors, etc. A wide range of features and methods for training sentiment classifiers for Twitter datasets have been researched in recent years with varying results. In this paper, we introduce a novel approach of adding semantics as additional features into the training set for sentiment analysis. For each extracted entity (e.g. iPhone) from tweets, we add its semantic concept (e.g. Apple product) as an additional feature, and measure the correlation of the representative concept with negative/positive sentiment. We apply this approach to predict sentiment for three different Twitter datasets. Our results show an average increase of F harmonic accuracy score for identifying both negative and positive sentiment of around 6.5% and 4.8% over the baselines of unigrams and part-of-speech features respectively. We also compare against an approach based on sentiment-bearing topic analysis, and find that semantic features produce better Recall and F score when classifying negative sentiment, and better Precision with lower Recall and F score in positive sentiment classification.

The impact of employees' customer orientation and service orientation behaviours on customers' service evaluation

In this paper the authors conceptualize and test the effects of service employees’ customer orientation and service orientation behaviors within an extended service evaluation model encompassing service quality, service encounter quality, perceived value and customer satisfaction. The context is 271 Indian retail customers. Data analysis incorporates confirmatory factor analysis and structural equation modeling. We find that Findings indicate that: 1) customer orientation is positively related to service orientation, customers’ perceptions of service encounter quality and service quality; 2) service orientation influences customers’ perceptions of service encounter quality and service quality; 3) customers’ perceptions of service encounter quality influence customers’ perceptions of service quality and customer satisfaction; 4) customers’ perceptions of service quality influence value perceptions; 5) service quality influences customer satisfaction; and 6) customer satisfaction influences customers’ behavioral intentions. The importance of these findings for practitioners and academics, research limitations and future research avenues are subsequently discussed.

The relative influence of organisational commitment and job-satisfaction on service quality of customer-contact employees in banking call centres

An inter-disciplinary approach is adopted to provide a deeper understanding of the human resource-service quality relationship. The paper tests the relationships organisational commitment and job satisfaction have with service quality of customer-contact employees. Hypotheses are constructed by reviewing literature in the areas of human resource management and services marketing. A study comprising 342 employees was conducted in four telephone call centres of a major UK retail bank. Investigates how different forms of organisational commitment and job satisfaction influence the service quality delivered by contact employees. Findings indicate that job satisfaction and organisational commitment of employees have a significant impact on service quality delivered. The affective component of commitment was found to be more important than job satisfaction in determining service quality of customer-contact employees.

Impact of online channel use on customer revenues and costs to serve:considering product portfolios and self-selection

Developing a strategy for online channels requires knowledge of the effects of customers' online use on their revenue and cost to serve, which ultimately influence customer profitability. The authors theoretically discuss and empirically examine these effects. An empirical study of retail banking customers reveals that online use improves customer profitability by increasing customer revenue and decreasing cost to serve. Moreover, the revenue effects of online use are substantially larger than the cost-to-serve effects, although the effects of online use on customer revenue and cost to serve vary by product portfolio. Self-selection effects also emerge and can be even greater than online use effects. Ignoring self-selection effects thus can lead to poor managerial decision-making.

The impact of the introduction and use of an informational website on offline customer buying behavior

Do customers increase or decrease their spending in response to the introduction of an informational website? To answer this question, this study considers the effects of the introduction and use of an informational website by a large national retailer on offline customer buying behavior. More specifically, we study a website's effects on the number of shopping trips and the amount spent per category per shopping trip. The model is calibrated through the estimation of a Poisson model (shopping trips) and a type-II tobit model (the amount spent per category per shopping trip), with effect parameters that vary across customers. For the focal retailer, an informational website creates more bad than good news; most website visitors engage in fewer shopping trips and spend less in all product categories. The authors also compare the characteristics of shoppers who exhibit negative website effects with those few shoppers who show positive effects and thus derive key implications for research and practice.

Analytics for customer engagement

In this article, we discuss the state of the art of models for customer engagement and the problems that are inherent to calibrating and implementing these models. The authors first provide an overview of the data available for customer analytics and discuss recent developments. Next, the authors discuss the models used for studying customer engagement, where they distinguish the following stages: customer acquisition, customer development, and customer retention. Finally, they discuss several organizational issues of analytics for customer engagement, which constitute barriers for introducing analytics for customer engagement.

Rewarding customer service? Using reward and recognition to deliver your customer service strategy

This report examines the role of reward and recognition in helping organisations deliver excellent customer service. It identifies the impact that reward and recognition practice has upon customer service and highlights which practices are more effective; shows the extent to which employees are satisfied with the rewards and recognition they receive; indicates those approaches to reward and recognition most associated with staff commitment; and looks at how other factors affect employee satisfaction with reward and recognition.

Joint sentiment/topic model for sentiment analysis

Sentiment analysis or opinion mining aims to use automated tools to detect subjective information such as opinions, attitudes, and feelings expressed in text. This paper proposes a novel probabilistic modeling framework based on Latent Dirichlet Allocation (LDA), called joint sentiment/topic model (JST), which detects sentiment and topic simultaneously from text. Unlike other machine learning approaches to sentiment classification which often require labeled corpora for classifier training, the proposed JST model is fully unsupervised. The model has been evaluated on the movie review dataset to classify the review sentiment polarity and minimum prior information have also been explored to further improve the sentiment classification accuracy. Preliminary experiments have shown promising results achieved by JST.

Understanding relationship marketing from the customer's perspective

DUE TO COPYRIGHT RESTRICTIONS ONLY AVAILABLE FOR CONSULTATION AT ASTON UNIVERSITY LIBRARY AND INFORMATION SERVICES WITH PRIOR ARRANGEMENT

Managing customer relationship management projects:the case of a large French telecommunication company

Customer relationship management (CRM) implementation projects reflect a growing conceptual shift from the traditional engineering view of projects. Such projects are complex and risky because they call for both organisational and technological changes. This requires effective project management across various phases of the implementation process. However, few empirical researches have dealt with these project management issues. The aim of this research is to investigate how a “project team” manages CRM implementation projects successfully, across the different phases of the implementation process. We conducted an in-depth case study of the “Firm-Clients Branch” of a large telecommunications company in France. The findings show that, to manage CRM implementation projects successfully, an integrated and balanced approach is required. This involves appropriate system selection, effective process re-engineering and further development of organizational structures. We highlight the need for a “technochange approach” to achieve successful organisational transition and effective CRM implementation. The study reveals that the project team plays a central role throughout the implementation phases. Furthermore the effectiveness of technochange depends on project team performance, technology efficiency and close coordination with stakeholders.

Does the importance of value, brand and relationship equity for customer loyalty differ between Eastern and Western cultures?

The use of the customer equity framework as a focal marketing strategy to increase customer loyalty has emerged as an important topic. Despite a growing number of investigations, previous studies are limited by their strong U.S. and European orientations. Research into Western consumers cannot necessarily predict the behaviour of Eastern consumers though. Therefore, this study investigates whether the link between customer equity drivers (value equity, brand equity and relationship equity) and loyalty intentions is sensitive to the cultural environment. A sample of 1553 Chinese and 1085 Dutch consumers in the banking and supermarket industries reveals that all three customer equity drivers exert a greater impact in Western than in Eastern cultures. This study also shows that Eastern consumers in general have higher loyalty intentions than Western consumers. © 2013.

Dynamic joint sentiment-topic model

Social media data are produced continuously by a large and uncontrolled number of users. The dynamic nature of such data requires the sentiment and topic analysis model to be also dynamically updated, capturing the most recent language use of sentiments and topics in text. We propose a dynamic Joint Sentiment-Topic model (dJST) which allows the detection and tracking of views of current and recurrent interests and shifts in topic and sentiment. Both topic and sentiment dynamics are captured by assuming that the current sentiment-topic-specific word distributions are generated according to the word distributions at previous epochs. We study three different ways of accounting for such dependency information: (1) Sliding window where the current sentiment-topic word distributions are dependent on the previous sentiment-topic-specific word distributions in the last S epochs; (2) skip model where history sentiment topic word distributions are considered by skipping some epochs in between; and (3) multiscale model where previous long- and shorttimescale distributions are taken into consideration. We derive efficient online inference procedures to sequentially update the model with newly arrived data and show the effectiveness of our proposed model on the Mozilla add-on reviews crawled between 2007 and 2011. © 2013 ACM 2157-6904/2013/12-ART5 $ 15.00.