Prediction of age, sentiment, and connectivity from social media text

Social media corpora, including the textual output of blogs, forums, and messaging applications, provide fertile ground for linguistic analysis material diverse in topic and style, and at Web scale. We investigate manifest properties of textual messages, including latent topics, psycholinguistic features, and author mood, of a large corpus of blog posts, to analyze the impact of age, emotion, and social connectivity. These properties are found to be significantly different across the examined cohorts, which suggest discriminative features for a number of useful classification tasks. We build binary classifiers for old versus young bloggers, social versus solo bloggers, and happy versus sad posts with high performance. Analysis of discriminative features shows that age turns upon choice of topic, whereas sentiment orientation is evidenced by linguistic style. Good prediction is achieved for social connectivity using topic and linguistic features, leaving tagged mood a modest role in all classifications.

Event extraction using behaviors of sentiment signals and burst structure in social media

Significant world events often cause the behavioral convergence of the expression of shared sentiment. This paper examines the use of the blogosphere as a framework to study user psychological behaviors, using their sentiment responses as a form of ‘sensor’ to infer real-world events of importance automatically. We formulate a novel temporal sentiment index function using quantitative measure of the valence value of bearing words in blog posts in which the set of affective bearing words is inspired from psychological research in emotion structure. The annual local minimum and maximum of the proposed sentiment signal function are utilized to extract significant events of the year and corresponding blog posts are further analyzed using topic modeling tools to understand their content. The paper then examines the correlation of topics discovered in relation to world news events reported by the mainstream news service provider, Cable News Network, and by using the Google search engine. Next, aiming at understanding sentiment at a finer granularity over time, we propose a stochastic burst detection model, extended from the work of Kleinberg, to work incrementally with stream data. The proposed model is then used to extract sentimental bursts occurring within a specific mood label (for example, a burst of observing ‘shocked’). The blog posts at those time indices are analyzed to extract topics, and these are compared to real-world news events. Our comprehensive set of experiments conducted on a large-scale set of 12 million posts from Livejournal shows that the proposed sentiment index function coincides well with significant world events while bursts in sentiment allow us to locate finer-grain external world events.

A sentiment-aware approach to community formation in social media

Participating in a community exemplifies the aspect of sharing, networking and interacting in a social media system. There has been extensive work on characterising on-line communities by their contents and tags using topic modelling tools. However, the role of sentiment and mood has not been studied. Arguably, mood is an integral feature of a text, and becomes more significant in the context of social media: two communities might discuss precisely the same topics, yet within an entirely different atmosphere. Such sentiment-related distinctions are important for many kinds of analysis and applications, such as community recommendation. We present a novel approach to identification of latent hyper-groups in social communities based on users’ sentiment. The results show that a sentiment-based approach can yield useful insights into community formation and metacommunities, having potential applications in, for example, mental health—by targeting support or surveillance to communities with negative mood—or in marketing—by targeting customer communities having the same sentiment on similar topics.

War and private sentiment in Australia during 1915

This article investigates the development of a total war mentality in Australia during the First World War. Through a study of private letters and diaries, it observes the much greater level of popular commitment to the war that emerged in the middle of 1915, and an increasing acceptance throughout that year that the expanding war had taken on a life of its own, and that it would not end suddenly or without tremendous sacrifice. By the end of 1915, Australians were showing ever greater levels of dedication to a war offering increasingly less sense of how long it might continue.

Enhanced Twitter sentiment analysis by using feature selection and combination

Tweet sentiment analysis is an important research topic. An accurate and timely analysis report could give good indications on the general public's opinions. After reviewing the current research, we identify the need of effective and efficient methods to conduct tweet sentiment analysis. This paper aims to achieve a high level of performance for classifying tweets with sentiment information. We propose a feasible solution which improves the level of accuracy with good time efficiency. Specifically, we develop a novel feature combination scheme which utilizes the sentiment lexicons and the extracted tweet unigrams of high information gain. We evaluate the performance of six popular machine learning classifiers among which the Naive Bayes Multinomial (NBM) classifier achieves the accuracy rate of 84.60% and takes a few minutes to complete classifying thousands of tweets.

Initial public offerings in Australia 1994 to 1999, recent evidence of underpricing and underperformance

This paper analyses whether financial and non financial characteristics of Australian initial public offerings (IPOs) can explain observed underpricing and long term underperformance over the period 1994 to 1999. A number of previous Australian studies have investigated initial day underpricing and longer term underperformance of IPOs and this study updates those papers. We find that initial day underpricing can in part be explained by market sentiment, forecast dividend per share yields, underwriter options and share options. Our longer term analysis supports the finding of previous studies in that IPOs on average, underperform the market in the first year following their listing.

Scientific proof that humans enjoy punishing wrongdoers: The implications for punishment and sentencing

Recent developments in brain science confirm that as a race we are in fact a punitive lot. Human beings actually derive pleasure from inflicting punishment on wrongdoers. We are wired in such a way that the part of our brain that reports pleasure is activated when we punish norm violators. This is even when punishment has no tangible or demonstrable benefits. However, we are not slaves lo our emotions. Another region of our brain 'kicks-in' if punishment becomes self-defeating, in that it conflicts with our other interests. The implications of this research for punishment theory and the practice of sentencing are discussed in this paper. The findings give qualified support to the theory known as intrinsic retributivism, but do not suggest it is the soundest theory of punishment. This is because we stop punishing when it comes at a cost to us. The good feeling that punishment invokes in punishers is another consequential consideration in favour of the utilitarian theory of punishment. However, it is not clear that the utilitarian calculus is necessarily affected by the findings. The main implication of the research findings relates to the relevance of public opinion to sentencing practice. The findings support the view that public sentiment, which seems to support increasingly tougher sanctions, can be curtailed of the public are informed that punishment comes of a cost to community.

The pricing of property trust IPOs in Australia

Following Brounen and Eichholtz (2002) this paper adds to the international literature investigating the underpricing of REIT initial public offerings (IPOs), with a study into Australian property trusts. This study finds that initial day returns can in part be explained by forecast profit distributions (or dividends) and the market sentiment towards property trusts from the date of the prospectus to the date of listing. There is some support for the “winners curse” explanation of underpricing with evidence that large investor or institutional involvement at the outset of the IPO also has some explanatory power.

Differences in underpricing returns between REIT IPOs and industrial company IPOs

This paper follows Chan, Stohs, andWang (2001), which argues that the underlying value of the real estate is not by itself the reason for the very substantial differences in underpricing returns between real estate investment trust (REIT) IPOs and industrial company IPOs.We use variables identified in previous studies that have helped explain the underpricing of industrial company IPOs to help explain the underpricing of property trust IPOs. We find that the prospectus forecast dividend yield is a critical variable in the valuation and hence underpricing of REIT IPOs compared to industrial company IPOs. The sentiment towards the market and whether or not the issue is underwritten also impact the underpricing of REITs but the impact is much less than on industrial company IPOs.

Anti-Americanism: recent sources

Anti-American sentiment has, it would seem, increased in the wake of the election of George Bush, jr. and the US’s response to the events of
September 11. Commentators within the US, and globally, have analysed the recent variants of a long-standing condition. Below is a review of recent coverage of the phenomenon. The entries for the print sources are divided into:

(1) analyses of anti-Americanism written in the US,
(2) accounts of international expressions of anti-Americanism.

Conversations with my mother

An intense and sometimes disturbing series of encounters between the filmmaker and his mother as they relive the traumatic years of his childhood and adolescence. Following the migration of the family to Australia from Holland in the difficult postwar years they had to grapple with problems of housing, social injustice and adjustment made more difficult by the father's mental illness. For the filmmaker 'the sentiment had to be uncompromisingly true' although he became aware that 'all film is fiction'.

Applying outcomes of lifelong learning to organisational achievement

The focus of this doctoral research study is making the most what a person knows and can do, as an outcome from their lifelong learning, so as to better contribute to organisational achievement. This has been motivated by a perceived gap in the extensive literature linking knowledge with organisational achievement. Whereas there is a rich body of literature addressing the meta-philosophies giving rise to the emergence of learning organisations there is, as yet, scant attention paid to the detail of planning and implementing action which would reveal individual/organisational opportunities of mutual advantage and motivate, and sustain, participation at the day-to-day level of the individual. It is in this space that this dissertation seeks to contribute by offering a mechanism for bringing the, hindsight informed, response “but that’s obvious” into the abiding explicit realm at the level of the individual. In moving beyond the obvious which is prone to be overlooked, the emphasis on “better” in the introductory sentence, is very deliberately made and has a link to awakening latent individual, and hence organisational, capabilities that would otherwise languish. The evolved LCM Model – a purposeful integration valuing the outcomes from lifelong learning (the L) with nurturing a culture supporting this outcome (the C) and with responsiveness to potentially diverse motivations (the M) – is a reflective device for bringing otherwise tacit, and latent, logic into the explicit realm of action. In the course of the development of the model, a number of supplementary models included in this dissertation have evolved from the research. They form a suite of devices which inform action and lead to making the most of what an individual knows and can do within the formal requirements of a job and within the informal influences of a frequently invisible community of practice. The initial inquiry drew upon the views and experiences of water industry engineering personnel and training facilitators associated with the contract cleaning and waste management industries. However, the major research occurred as an Emergency Management Australia (EMA) project with the Country Fire Authority (CFA) as the host organisation. This EMA/CFA research project explored the influence of making the most of what a CFA volunteer knows and can do upon retention of that volunteer. In its aggregate, across the CFA volunteer body, retention is a critical community safety objective. A qualitative research, ethnographic in character, approach was adopted. Data was collected through interviews, workshops and outcomes from attempts at action research projects. Following an initial thirteen month scoping study including respondents other than from the CFA, the research study moved into an exploration of the efficacy of an indicative model with four contextual foci – i.e. the manner of welcoming new members to the CFA, embracing training, strengthening brigade sustainability and leadership. Interestingly, the research environment which forced a truncated implementation of action research projects was, in itself, an informing experience indicative of inhibitors to making the most of what people know and can do. Competition for interest, time and commitment were factors governing the manner in which CFA respondents could be called upon to explore the efficacy of the model, and were a harbinger of the influences shaping the more general environment of drawing upon what CFA volunteers know and can do. Subsequent to the development of the indicative model, a further 16 month period was utilised in the ethnographic exploration of the relevance of the model within the CFA as the host organisation. As a consequence, the model is a more fully developed tool (framework) to aid reflection, planning and action. Importantly, the later phase of the research study has, through application of the model to specific goals within the CFA, yielded operational insight into its effective use, and in which activity systems have an important place. The model – now confidently styled as the LCM Model – has three elements that when enmeshed strengthen the likelihood of organisational achievement ; and the degree of this meshing, as relevant to the target outcome, determines the strength of outcome. i.e. - • Valuing outcomes from learning: When a person recognises and values (appropriately to achievement by the organisation) what they know and can do, and associated others recognise and value what this person knows and can do, then there is increased likelihood of these outcomes from learning being applied to organisational achievement. • Valuing a culture that is conducive to learning: When a person, and associated others, are further developing and drawing upon what they know and can do within the context of a culture that is conducive to learning, then there is increased likelihood that outcomes from learning will be applied to organisational achievement. • Valuing motivation of the individual: When a person’s motivation to apply what they know and can do is valued by them, and associated others, as appropriate to organisational achievement then there is increased likelihood that appropriately drawing upon outcomes from learning will occur. Activity theory was employed as a device to scope and explore understanding of the issues as they emerged in the course of the research study. Viewing the data through the prism of activity theory led not only to the development of the LCM Model but also to an enhanced understanding of the role of leadership as a foundation for acting upon the model. Both formal and informal leadership were found to be germane in asserting influence on empowering engagement with learning and drawing upon its outcomes. It is apparent that a “leaderful organisation”, as postulated by Raelin (2003), is an environment which supports drawing upon the LCM model; and it may be the case that the act of drawing upon the model will move a narrowly leadership focused organisation toward leaderful attributes. As foreshadowed at the beginning of this synopsis, nurturing individual and organisational capability is the guiding mantra for this dissertation - “Capability embraces competence but is also forward-looking, concerned with the realisation of potential” (Stephenson 1998, p. 3). Although the inquiry focussed upon a need for CFA volunteer retention, it began with a broader investigation as part of the scoping foundation and the expanded usefulness of the LCM Model invites further investigation. The dissertation concludes with the encapsulating sentiment that “You have really got to want to”. With this predisposition in mind, this dissertation contributes to knowledge through the development and discussion of the LCM model as a reflective device informing transformative learning (Mezirow and Associates 1990). A leaderful environment (Raelin 2003) aids transformative learning – accruing to the individual and the organisation - through engendering and maintaining making the most of knowledge and skill – motivating and sustaining “the will”. The outcomes from this research study are a strong assertion that wanting to make the most of what is known and can be done is a hallmark of capability. Accordingly, this dissertation is a contribution to the “how” of strengthening the capability, and the commitment to applying that capability, of an individual and an organisation.

Australian and commonwealth republicanism

When republics, beginning with India in 1949, were first admitted to the Commonwealth of Nations, Australia remained strongly attached to the Crown and the King’s (later the Queen’s) role as Head of the Commonwealth. Indeed, many Australians had seen a shared Crown as axiomatic, and a symbol of Commonwealth unity. Despite bursts of republicanism in Australia during the 19th and 20th centuries, it was not until the 1990s that a republic appeared likely. One historic driver of anti-British Australian republicanism has been the Irish heritage of many Australians. As republicanism grew, it was important that Australia could remain in the Commonwealth as a republic. The past decade has seen a stronger sentiment in Australia than in the other ‘old Dominions’—New Zealand and Canada—that national independence and identity require the symbol of a home-grown head of state, rather than one seen as British. The growth of republicanism in such countries, and in Britain itself, would be likely to encourage republicanism in Australia. Australia’s republican majority has been frustrated by its inability to agree on a model for parliamentary selection or direct election of the president. No Commonwealth country provides a model which Australians find compelling.