While most work in sentiment analysis in the financial domain has focused on the use of content from traditional finance news, in this work we concentrate on more subjective sources of information, blogs. We aim to automatically determine the sentiment of financial bloggers towards companies and their stocks. To do this we develop a corpus of financial blogs, annotated with polarity of sentiment with respect to a number of companies. We conduct an analysis of the annotated corpus, from which we show there is a significant level of topic shift within this collection, and also illustrate the difficulty that human annotators have when annotating certain sentiment categories. To deal with the problem of topic shift within blog articles, we propose text extraction techniques to create topic-specific sub-documents, which we use to train a sentiment classifier. We show that such approaches provide a substantial improvement over full documentclassification and that word-based approaches perform better than sentence-based or paragraph-based approaches.

In this paper we present the results of our work on the analysis of multi-modal data for video Information Retrieval, where we exploit the properties of this data for query-time, automatic generation of weights for multi-modal data fusion. Through empirical testing we have observed that for a given topic, a high performing feature, that is one which achieves high relevance, will have a different distribution of document scores when compared against those that do not perform as well. These observations form the basis for our initial fusion model, which generates weights based on these properties, without the need for prior training. Our model can be used to not only combine feature data, but to also combine the results of multiple example query images and apply weights to these. Our analysis and experiments were conducted on the TRECVid 2004 and 2005 collections, making use of multiple MPEG-7 low-level features and automatic speech recognition (ASR) transcripts. Results achieved from our model achieve performance on a par with that of 'oracle' determined weights, and demonstrate the applicability of our model whilst advancing the case for further investigation of score distributions.

Steady progress in the field of multimedia information retrieval (MMIR) promises a useful set of tools that could provide new usage scenarios and features to enhance the user experience in today's digital media applications. In the interactive TV domain, the simplicity of interaction is more crucial than in any other digital media domain and ultimately determines the success or otherwise of any new applications. Thus when integrating emerging tools like MMIR into interactive TV, the increase in interface complexity and sophistication resulting from these features can easily reduce its actual usability. In this paper we describe a design strategy we developed as a result of our efforts in balancing the power of emerging multimedia information retrieval techniques and maintaining the simplicity of the interface in interactive TV. By providing multiple levels of interface sophistication in increasing order as a viewer repeatedly presses the same button on their remote control, we provide a layered interface that can accommodate viewers requiring varying degrees of power and simplicity. A series of screen shots from the system we have actually developed and built illustrates how this is achieved.

Abstract-In this paper we describe how content-based analysis techniques can be used to provide much greater functionality to the users of an interactive TV (iTV) device. We describe several content-based multimedia analysis techniques and how some of these can be exploited in the iTV domain, resulting in the provision of a set of powerful functions for iTV users. To validate our ideas, we introduce an iTV application we developed which incorporates some of these techniques into a simple set of user features, in order to demonstrate the usefulness of content-based techniques for iTV. The contribution of this paper is not to provide an in-depth discussion on each of the individual content-based techniques, but rather to show how many of these powerful technologies can be incorporated into an interactive TV system.