Learning user and product distributed representations using a sequence model for sentiment analysis

In product reviews, it is observed that the distribution of polarity ratings over reviews written by different users or evaluated based on different products are often skewed in the real world. As such, incorporating user and product information would be helpful for the task of sentiment classification of reviews. However, existing approaches ignored the temporal nature of reviews posted by the same user or evaluated on the same product. We argue that the temporal relations of reviews might be potentially useful for learning user and product embedding and thus propose employing a sequence model to embed these temporal relations into user and product representations so as to improve the performance of document-level sentiment analysis. Specifically, we first learn a distributed representation of each review by a one-dimensional convolutional neural network. Then, taking these representations as pretrained vectors, we use a recurrent neural network with gated recurrent units to learn distributed representations of users and products. Finally, we feed the user, product and review representations into a machine learning classifier for sentiment classification. Our approach has been evaluated on three large-scale review datasets from the IMDB and Yelp. Experimental results show that: (1) sequence modeling for the purposes of distributed user and product representation learning can improve the performance of document-level sentiment classification; (2) the proposed approach achieves state-of-The-Art results on these benchmark datasets.

Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN

Different types of sentences express sentiment in very different ways. Traditional sentence-level sentiment classification research focuses on one-technique-fits-all solution or only centers on one special type of sentences. In this paper, we propose a divide-and-conquer approach which first classifies sentences into different types, then performs sentiment analysis separately on sentences from each type. Specifically, we find that sentences tend to be more complex if they contain more sentiment targets. Thus, we propose to first apply a neural network based sequence model to classify opinionated sentences into three types according to the number of targets appeared in a sentence. Each group of sentences is then fed into a one-dimensional convolutional neural network separately for sentiment classification. Our approach has been evaluated on four sentiment classification datasets and compared with a wide range of baselines. Experimental results show that: (1) sentence type classification can improve the performance of sentence-level sentiment analysis; (2) the proposed approach achieves state-of-the-art results on several benchmarking datasets.

A Bayesian modeling approach to multi-dimensional sentiment distributions prediction

Sentiment analysis has long focused on binary classification of text as either positive or negative. There has been few work on mapping sentiments or emotions into multiple dimensions. This paper studies a Bayesian modeling approach to multi-class sentiment classification and multidimensional sentiment distributions prediction. It proposes effective mechanisms to incorporate supervised information such as labeled feature constraints and document-level sentiment distributions derived from the training data into model learning. We have evaluated our approach on the datasets collected from the confession section of the Experience Project website where people share their life experiences and personal stories. Our results show that using the latent representation of the training documents derived from our approach as features to build a maximum entropy classifier outperforms other approaches on multi-class sentiment classification. In the more difficult task of multi-dimensional sentiment distributions prediction, our approach gives superior performance compared to a few competitive baselines. © 2012 ACM.

Social media and sentiment in bioenergy consultation

he push to widen participation in public consultation suggests social media as an additional mechanism through which to engage the public. Bioenergy companies need to build their capacity to communicate in these new media and to monitor the attitudes of the public and opposition organisations towards energy development projects. Design/methodology/approach This short paper outlines the planning issues bioenergy developments face and the main methods of communication used in the public consultation process in the UK. The potential role of social media in communication with stakeholders is identified. The capacity of sentiment analysis to mine opinions from social media is summarised, and illustrated using a sample of tweets containing the term ‘bioenergy’ Findings Social media have the potential to improve information flows between stakeholders and developers. Sentiment analysis is a viable Purpose The push to widen participation in public consultation suggests social media as an additional mechanism through which to engage the public. Bioenergy companies need to build their capacity to communicate in these new media and to monitor the attitudes of the public and opposition organisations towards energy development projects. Design/methodology/approach This short paper outlines the planning issues bioenergy developments face and the main methods of communication used in the public consultation process in the UK. The potential role of social media in communication with stakeholders is identified. The capacity of sentiment analysis to mine opinions from social media is summarised, and illustrated using a sample of tweets containing the term ‘bioenergy’ Findings Social media have the potential to improve information flows between stakeholders and developers. Sentiment analysis is a viable methodology, which bioenergy companies should be using to measure public opinion in the consultation process. Preliminary analysis shows promising results. Research limitations/implications Analysis is preliminary and based on a small dataset. It is intended only to illustrate the potential of sentiment analysis and not to draw general conclusions about the bioenergy sector. Originality/value Opinion mining, though established in marketing and political analysis, is not yet systematically applied as a planning consultation tool. This is a missed opportunity.

Challenging masculinity in CSR disclosures:silencing of women’s voices in Tanzania’s mining industry

This paper presents a feminist analysis of corporate social responsibility (CSR) in a male-dominated industry within a developing country context. It seeks to raise awareness of the silencing of women’s voices in CSR reports produced by mining companies in Tanzania. Tanzania is one of the poorest countries in Africa, and women are often marginalised in employment and social policy considerations. Drawing on work by Hélène Cixous, a post-structuralist/radical feminist scholar, the paper challenges the masculinity of CSR discourses that have repeatedly masked the voices and concerns of ‘other’ marginalised social groups, notably women. Using interpretative ethnographic case studies, the paper provides much-needed empirical evidence to show how gender imbalances remain prevalent in the Tanzanian mining sector. This evidence draws attention to the dynamics faced by many women working in or living around mining areas in Tanzania. The paper argues that CSR, a discourse enmeshed with the patriarchal logic of the contemporary capitalist system, is entangled with tensions, class conflicts and struggles which need to be unpacked and acknowledged. The paper considers the possibility of policy reforms in order to promote gender balance in the Tanzanian mining sector and create a platform for women’s concerns to be voiced.

Making sense of microposts (#MSM2013) concept extraction challenge

Microposts are small fragments of social media content that have been published using a lightweight paradigm (e.g. Tweets, Facebook likes, foursquare check-ins). Microposts have been used for a variety of applications (e.g., sentiment analysis, opinion mining, trend analysis), by gleaning useful information, often using third-party concept extraction tools. There has been very large uptake of such tools in the last few years, along with the creation and adoption of new methods for concept extraction. However, the evaluation of such efforts has been largely consigned to document corpora (e.g. news articles), questioning the suitability of concept extraction tools and methods for Micropost data. This report describes the Making Sense of Microposts Workshop (#MSM2013) Concept Extraction Challenge, hosted in conjunction with the 2013 World Wide Web conference (WWW'13). The Challenge dataset comprised a manually annotated training corpus of Microposts and an unlabelled test corpus. Participants were set the task of engineering a concept extraction system for a defined set of concepts. Out of a total of 22 complete submissions 13 were accepted for presentation at the workshop; the submissions covered methods ranging from sequence mining algorithms for attribute extraction to part-of-speech tagging for Micropost cleaning and rule-based and discriminative models for token classification. In this report we describe the evaluation process and explain the performance of different approaches in different contexts.

Mainstream media behavior analysis on Twitter:a case study on UK general election

With the development of social media tools such as Facebook and Twitter, mainstream media organizations including newspapers and TV media have played an active role in engaging with their audience and strengthening their influence on the recently emerged platforms. In this paper, we analyze the behavior of mainstream media on Twitter and study how they exert their influence to shape public opinion during the UK's 2010 General Election. We first propose an empirical measure to quantify mainstream media bias based on sentiment analysis and show that it correlates better with the actual political bias in the UK media than the pure quantitative measures based on media coverage of various political parties. We then compare the information diffusion patterns from different categories of sources. We found that while mainstream media is good at seeding prominent information cascades, its role in shaping public opinion is being challenged by journalists since tweets from them are more likely to be retweeted and they spread faster and have longer lifespan compared to tweets from mainstream media. Moreover, the political bias of the journalists is a good indicator of the actual election results. Copyright 2013 ACM.

Capturing strategy as an everyday practice:quantitative methodologies for analysing practice data

Few works address methodological issues of how to conduct strategy-as-practice research and even fewer focus on how to analyse the subsequent data in ways that illuminate strategy as an everyday, social practice. We address this gap by proposing a quantitative method for analysing observational data, which can complement more traditional qualitative methodologies. We propose that rigorous but context-sensitive coding of transcripts can render everyday practice analysable statistically. Such statistical analysis provides a means for analytically representing patterns and shifts within the mundane, repetitive elements through which practice is accomplished. We call this approach the Event Database (EDB) and it consists of five basic coding categories that help us capture the stream of practice. Indexing codes help to index or categorise the data, in order to give context and offer some basic information about the event under discussion. Indexing codes are descriptive codes, which allow us to catalogue and classify events according to their assigned characteristics. Content codes are to do with the qualitative nature of the event; this is the essence of the event. It is a description that helps to inform judgements about the phenomenon. Nature codes help us distinguish between discursive and tangible events. We include this code to acknowledge that some events differ qualitatively from other events. Type events are codes abstracted from the data in order to help us classify events based on their description or nature. This involves significantly more judgement than the index codes but consequently is also more meaningful. Dynamics codes help us capture some of the movement or fluidity of events. This category has been included to let us capture the flow of activity over time.

Constructions of active womanhood and new femininities: from a feminist linguistic perspective, is Sex and the City a modernist or a post-modernist TV text?

Sex and the City has been the subject of close scrutiny within feminist scholarship in terms of whether it is considered to be a reactionary or progressive text. While this debate is valuable within a modernist feminist paradigm, it makes less sense from a post-modernist feminist perspective. Alternately using semiotic and feminist post-structuralist methods of textual analysis, this paper shows that Sex and the City can be viewed as reactionary according to a modernist reading, but is altogether more challenging and complex according to a post-modernist reading.

A configuration framework of manufacturing strategy paradigm in state owned enterprises within China's petrochemical industry:building context-embedded configurations for the paradigm of strategic manufacturing management from exploratory case studies

The role of the production system as a key determinant of competitive performance of business operations- has long been the subject of industrial organization research, even predating the .explicit conceptua1isation of manufacturing, strategy in the literature. Particular emergent production issues such as the globalisation of production, global supply chain management, management of integrated manufacturing and a growing e~busjness environment are expected to critically influence the overall competitive performance and therefore the strategic success of the organization. More than ever, there is a critical need to configure and improve production system and operations competence in a strategic way so as to contribute to the long-term competitiveness of the organization. In order to operate competitively and profitably, manufacturing companies, no matter how well managed, all need a long-term 'strategic direction' for the development of operations competence in order to consistently produce more market value with less cost towards a leadership position. As to the long-term competitiveness, it is more important to establish a dynamic 'strategic perspective' for continuous operational improvements in pursuit of this direction, as well as ongoing reviews of the direction in relation to the overall operating context. However, it also clear that the 'existing paradigm of manufacturing strategy development' is incapable of adequately responding to the increasing complexities and variations of contemporary business operations. This has been factually reflected as many manufacturing companies are finding that methodologies advocated in the existing paradigm for developing manufacturing strategy have very limited scale and scope for contextual contingency in empirical application. More importantly, there has also emerged a deficiency in the multidimensional and integrative profile from a theoretical perspective when operationalising the underlying concept of strategic manufacturing management established in the literature. The point of departure for this study was a recognition of such contextual and unitary limitations in the existing paradigm of manufacturing strategy development when applied to contemporary industrial organizations in general, and Chinese State Owned Enterprises (SOEs) in particular. As China gradually becomes integrated into the world economy, the relevance of Western management theory and its paradigm becomes a practical matter as much as a theoretical issue. Since China markedly differs from Western countries in terms of culture, society, and political and economic systems, it presents promising grounds to test and refine existing management theories and paradigms with greater contextual contingency and wider theoretical perspective. Under China's ongoing programmes of SOE reform, there has been an increased recognition that strategy development is the very essence of the management task for managers of manufacturing companies in the same way as it is for their counterparts in Western economies. However, the Western paradigm often displays a rather naive and unitary perspective of the nature of strategic management decision-making, one which largely overlooks context-embedded factors and social/political influences on the development of manufacturing strategy. This thesis studies the successful experiences of developing manufacturing strategy from five high-performing large-scale SOEs within China’s petrochemical industry. China’s petrochemical industry constitutes a basic heavy industrial sector, which has always been a strategic focus for reform and development by the Chinese government. Using a confirmation approach, the study has focused on exploring and conceptualising the empirical paradigm of manufacturing strategy development practiced by management. That is examining the ‘empirical specifics’ and surfacing the ‘managerial perceptions’ of content configuration, context of consideration, and process organization for developing a manufacturing strategy during the practice. The research investigation adopts a qualitative exploratory case study methodology with a semi-structural front-end research design. Data collection follows a longitudinal and multiple-case design and triangulates case evidence from sources including qualitative interviews, direct observation, and a search of documentations and archival records. Data analysis follows an investigative progression from a within-case preliminary interpretation of facts to a cross-case search for patterns through theoretical comparison and analytical generalization. The underlying conceptions in both the literature of manufacturing strategy and related studies in business strategy were used to develop theoretical framework and analytical templates applied during data collection and analysis. The thesis makes both empirical and theoretical contributions to our understanding of 'contemporary management paradigm of manufacturing strategy development'. First, it provides a valuable contextual contingency of the 'subject' using the business setting of China's SOEs in petrochemical industry. This has been unpacked into empirical configurations developed for its context of consideration, its content and process respectively. Of special note, a lean paradigm of business operations and production management discovered at case companies has significant implications as an emerging alternative for high-volume capital intensive state manufacturing in China. Second, it provides a multidimensional and integrative theoretical profile of the 'subject' based upon managerial perspectives conceptualised at case companies when operationalising manufacturing strategy. This has been unpacked into conceptual frameworks developed for its context of consideration, its content constructs, and its process patterns respectively. Notably, a synergies perspective towards the operating context, competitive priorities and competence development of business operations and production management has significant implications for implementing a lean manufacturing paradigm. As a whole, in so doing, the thesis established a theoretical platform for future refinement and development of context-specific methodologies for developing manufacturing strategy.

Translation strategies and cross-cultural contraints : a case study of the translation of advertising texts

Since the transfer of a message between two cultures very frequently takes place through the medium of a written text qua communicative event, it would seem useful to attempt to ascertain whether there is any kind of pattern in the use of strategies for the effective interlingual transfer of this message. Awareness of potentially successful strategies, within the constraints of context, text type, intended TL function and TL reader profile will enhance quality and cost-effectiveness (time, effort, financial costs) in the production of the target text. Through contrastive analysis of pairs of advertising texts, SL and TL, French and English, this study will attempt to identify the nature of some recurring choices made by different translators in the attempt to recreate ST information in the TL in such a manner as to reproduce as closely as possible the informative, persuasive and affective functions of the text as advertising material. Whilst recurrence may be seen to be significant in terms of illustrating tendencies with regard to the solution of problems of translation, this would not necessarily be taken as confirmation of the existence of pre-determined or prescriptive rules. These tendencies could, however, be taken as a guide to potential solutions to certain kinds of context-bound and text-type specific problem. Analysis of translated text-pairs taken from the field of advertising should produce examples of constraints posed by the need to select the content, tone and form of the Target Text, in order to ensure maximum efficacy of persuasive effect and to ensure the desired outcome, as determined by the Source Text function. When evaluating the success of a translated advertising text, constraints could be defined in terms of the culture-specific references or assumptions on which a Source Text may build in order to achieve its intended communicative function within the target community.

Learning task specific distributed paragraph representations using a 2-tier convolutional neural network

We introduce a type of 2-tier convolutional neural network model for learning distributed paragraph representations for a special task (e.g. paragraph or short document level sentiment analysis and text topic categorization). We decompose the paragraph semantics into 3 cascaded constitutes: word representation, sentence composition and document composition. Specifically, we learn distributed word representations by a continuous bag-of-words model from a large unstructured text corpus. Then, using these word representations as pre-trained vectors, distributed task specific sentence representations are learned from a sentence level corpus with task-specific labels by the first tier of our model. Using these sentence representations as distributed paragraph representation vectors, distributed paragraph representations are learned from a paragraph-level corpus by the second tier of our model. It is evaluated on DBpedia ontology classification dataset and Amazon review dataset. Empirical results show the effectiveness of our proposed learning model for generating distributed paragraph representations.

Capturing strategy as an everyday practice:quantitative methodologies for analysing practice data

Few works address methodological issues of how to conduct strategy-as-practice research and even fewer focus on how to analyse the subsequent data in ways that illuminate strategy as an everyday, social practice. We address this gap by proposing a quantitative method for analysing observational data, which can complement more traditional qualitative methodologies. We propose that rigorous but context-sensitive coding of transcripts can render everyday practice analysable statistically. Such statistical analysis provides a means for analytically representing patterns and shifts within the mundane, repetitive elements through which practice is accomplished. We call this approach the Event Database (EDB) and it consists of five basic coding categories that help us capture the stream of practice. Indexing codes help to index or categorise the data, in order to give context and offer some basic information about the event under discussion. Indexing codes are descriptive codes, which allow us to catalogue and classify events according to their assigned characteristics. Content codes are to do with the qualitative nature of the event; this is the essence of the event. It is a description that helps to inform judgements about the phenomenon. Nature codes help us distinguish between discursive and tangible events. We include this code to acknowledge that some events differ qualitatively from other events. Type events are codes abstracted from the data in order to help us classify events based on their description or nature. This involves significantly more judgement than the index codes but consequently is also more meaningful. Dynamics codes help us capture some of the movement or fluidity of events. This category has been included to let us capture the flow of activity over time.