No description available.

No description available.

Service oriented computing is crucial to a large and growing number of computational undertakings. Central to its approach are the open and network-accessible services provided by many different organisations, and which in turn enable the easy creation of composite workflows. This leads to an environment containing many thousands of services, in which a programmer or automated composition system must discover and select services appropriate for the task at hand. This discovery and selection process is known as matchmaking. Prior work in the field has conceived the problem as one of sufficiently describing individual services using formal, symbolic knowledge representation languages. We review the prior work, and present arguments for why it is optimistic to assume that this approach will be adequate by itself. With these issues in mind, we examine how, by reformulating the task and giving the matchmaker a record of prior service performance, we can alleviate some of the problems. Using two formalisms—the incidence calculus and the lightweight coordination calculus—along with algorithms inspired by information retrieval techniques, we evolve a series of simple matchmaking agents that learn from experience how to select those services which performed well in the past, while making minimal demands on the service users. We extend this mechanism to the overlooked case of matchmaking in workflows using multiple services, selecting groups of services known to inter-operate well. We examine the performance of such matchmakers in possible future services environments, and discuss issues in applying such techniques in large-scale deployments.

When a signal is decomposed into frequency bands, demodulated into modulator and carrier pairs, and portrayed in a carrier frequency-versus modulator frequency domain, significant information may be automatically observed about the signal. We refer to this domain as the modulation spectral domain. The modulation spectrum is referred to as a windowed Fourier transform across time that produces an acoustic frequency versus modulation frequency representation of a signal. Previously, frameworks incorporating the discrete short-time modulation transform (DSTMT) and modulation spectrum have been designed mostly for filtering of speech signals. This modulation spectral domain is rarely, if ever, discussed in typical signal processing courses today, and we believe its current associated tools and applications are somewhat limited. We seek to revisit this domain to uncover more intuition, develop new concepts to extend its capabilities, and increase its applications, especially in the area of music data mining. A recent interest has risen in using modulation spectral features, which are features in the modulation spectral domain, for music data mining. The field of music data mining, also known as music information retrieval (MIR), has been rapidly developing over the past decade or so. One reason for this development is the aim to develop frameworks leveraging the particular characteristics of music signals instead of simply copying methods previously applied to its speech-centered predecessors, such as speech recognition, speech synthesis, and speaker identification. This research seeks to broaden the perspective and use of an existing modulation filterbank framework by exploiting modulation features well suited for music signals. The objective of this thesis is to develop a framework for extracting modulation spectral features from music and other signals. The purpose of extracting features from these signals is to perform data mining tasks, such as unsupervised source identification, unsupervised source separation, and audio synthesis. More specifically, this research emphasizes the following: the usefulness of the DSTMT and the modulation spectrum for music data mining tasks; a new approach to unsupervised source identification using modulation spectral features; a new approach to unsupervised source separation; a newly introduced analysis of FM features in an AM-dominated modulation spectra; and other applications.

Social media collections are becoming increasingly important in the everyday life of Internet users. Recent statistics show that sites hosting social media and community-generated content account for five of the top ten most visited websites in the United States [4], are visited regularly by a broad cross-section of Internet users [61, 67, 115] and host an enormous quantity of information [119, 48, 9]. The increasing importance and size of these collections requires that information retrieval systems pay special attention to these collections, and in particular pay attention to those aspects of social media collections that set them apart from the general web. Social media collections are interesting and challenging from the perspective of information retrieval systems. These collections are dynamic, with content being constantly added, removed and modified. These collections are time-sensitive, with the most recently added content often viewed as the most significant. These collections are richly structured, with authorship information, often threading structure and higher-level topical classifications. Although this type of collection structure is frequently critical for comprehension, it is rarely exploited in retrieval algorithms. This thesis investigates the hypothesis that we can improve retrieval performance in these collections by leveraging this type of structure. To evaluate this hypothesis, we present an exploration of search in several social media collections: blogs and online forums. We demonstrate the utility of leveraging collection structure in three different retrieval tasks: blog post search, blog feed search, and forum thread search. The techniques explored throughout these experiments include evaluating the representation granularity of collections of documents, and methods to incorporate content an author has written throughout the collection. Our results show that, although the retrieval tasks and techniques to leverage this type of collection structure are varied, in many cases substantial and significant retrieval quality improvements can be realized by leveraging this collection structure.

Information Retrieval, as a field, has long subscribed to an orthodox evaluation approach known as the Cranfield paradigm. This approach and the assumptions that underpin it have been essential to building the traditional search engine infrastructure that drives today’s modern information economy. In order to build the information economy of tomorrow, however, we must be prepared to reexamine these assumptions and create new, more sophisticated standards of evaluation to match the more complex information retrieval systems on the horizon. In this thesis, we begin this introspective process and launch our own evaluation method for one of these complex IR systems, entity-relationship recommenders. We will begin building a new user model adapted to the needs of a different user experience. To support these endeavors, we will also conduct a study with a mockup of our complex system to collect real behavior data and evaluation results. By the end of this work, we shall present a new evaluative approach for one kind of entity-relationship system and point the way for other advanced systems to come.

An individual’s ability to locate the information they desire grows more slowly than the rate at which new information becomes available. Customers are constantly confronted with situations in which they have many options to choose from and need assistance exploring or narrowing down the possibilities. Recommender systems are one tool to help bridge this gap. There are various mechanisms being employed to create recommender systems, but the most common systems fall into two main classes: content-based and collaborative filtering systems. Content-based recommender systems match the textual information of a particular product with the textual information representing the interests of a customer. Collaborative filtering systems use patterns in customer ratings to make recommendations. Both types of recommender systems require significant data resources in the form of a customer’s ratings and product features; hence they are not able to generate high quality recommendations. Hybrid mechanisms have been used by researchers to improve the performance of recommender systems where one can integrate more than one mechanism to overcome the drawbacks of an individual system. The hybrid approach proposed in this thesis is the integration of content and context-based with collaborative filtering, since these are the most successful and widely used mechanisms. This proposed approach will look into the integration of content and context data with rating data using a different mechanism that mainly focuses on boosting a customer’s trust in the recommender system. Researchers have been trying to improve system performance using hybrid approaches, but research is lacking on providing justifications for recommended products. Hence, the proposed approach will mainly focus on providing justifications for recommended products as this plays a crucial role in obtaining the satisfaction and trust of customers. A product’s features and a customer’s context attributes are used to provide justifications. In addition to this, the presentation mechanism needs to be very effective as it has been observed that customers trust more in a system when there are explanations on how the recommended products have been computed and presented. Finally, this proposed recommender system will allow the customer to interact with it in various ways to provide feedback on the recommendations and justifications. Overall, this integration will be very useful in achieving a stronger correlation between the customers and products. Experimental results clearly showed that the majority of the participants prefer to have recommendations with their justifications and they received valuable recommendations on which they could trust.

Many advantages of the Internetâ ease of access, limited regulation, vast potential audience, and fast flow of informationâ have turned it into the most popular way to communicate and exchange ideas. Criminal and terrorist groups also use these advantages to turn the Internet into their new play/battle fields to conduct their illegal/terror activities. There are millions of Web sites in different languages on the Internet, but the lack of foreign language search engines makes it impossible to analyze foreign language Web sites efficiently. This thesis will enhance an open source Web crawler with Arabic search capability, thus improving an existing social networking tool to perform page correlation and analysis of Arabic Web sites. A social networking tool with Arabic search capabilities could become a valuable tool for the intelligence community. Its page correlation and analysis results could be used to collect open source intelligence and build a network of Web sites that are related to terrorist or criminal activities.

This exploratory research investigates the factors that might influence a specific group of users’ mental models of a Web search engine, Google, as measured in the dimension of completeness. A modified mental model completeness scale (MMCS) was developed based on Borgman’s, Dimitroff s, and Saxon’s models, encompassing the perception of (1) the nature of the Web search engine, (2) searching features of the Web search engine, and (3) the interaction between the searcher and the Web search engine. With this scale, a participant’s mental model completeness level was determined by how many components of the first two parts of the scale were described and which level of interaction between the participant and Google was revealed during the searches. The choice of the factors was based on the previous studies on individual differences among information seekers, including user’s search experience, cognitive style, learning style, technical aptitudes, training received, discipline, and gender. Sixteen Ph.D. students whose first language is English participated in the research. Individual semi-structured interviews were conducted to determine the students’ mental model completeness level (MMCL) as well as their search experience, training received, discipline and gender. Direct observation technique was employed to observe students’ actual interactions with Google. Standard tests were administered to determine the students’ cognitive styles, learning styles and technical aptitudes. / Cette recherche préliminaire examine les facteurs qui peuvent influencer les modèles mentaux d’un groupe spécifique d’utilisateurs d’un moteur de recherche sur le Web: Google, mesurés selon l’étendue de leur réussite.Une échelle de cette réussite en suivant un modèle mental a été constituée en adaptant les modèles présentés par Borgman, Dimitroff et Saxon, incluant la perception (1) de la nature du moteur de recherche sur le Web, (2) des caractéristiques de la recherche propres à ce moteur, (3) de l’interaction entre le chercheur et le moteur de recherche. A l’aide de cette échelle, le niveau de réussite par un sujet donné utilisant un modèle mental a été déterminé en fonction du nombre de composantes des deux premières parties de l’échelle décrites et du niveau d’interaction entre le sujet et le moteur Google, tel que révélé par ses recherches. Le choix des facteurs a été fondé sur des études précédentes portant sur les différences individuelles entre les chercheurs d’information, comprenant le degré d’expérience d’une telle recherche par l’utilisateur, son style cognitif, son style d’apprentissage, ses aptitudes techniques, la formation reçue, la discipline et le sexe. Seize étudiants en doctorat ayant l’anglais comme première langue ont participé à cette étude. Des entretiens individuels semi-dirigés ont permis de déterminer le niveau de réussite des étudiants suivant leur modèle mental, ainsi que leur expérience de la recherche, la formation reçue, la discipline et le sexe. Une observation technique directe a été utilisée pour observer l’interaction réelle des étudiants avec Google. Des tests standardisés ont été administrés pour déterminer le style cognitif des étudiants, leur style d’apprentissage et leurs aptitudes techniques. fr

Increasingly, workplace information seeking takes place in digital information environments and is reliant upon search systems. Existing systems are designed to retrieve information that is relevant to the query, but are not capable of identifying information that is well-suited to the context and situation of a search. This is a problem for professionals who often are searching for a small amount of useful information that can be applied to a problem or task, and have limited time to browse through large sets of results. This inability of search systems to discriminate between relevant and useful documents is one of the core problems in information retrieval. In this dissertation, I address this problem by studying the role that contextual factors play in determining how a group of professionals searches for and selects information. The central question concerns the nature of the relationships between these contextual factors, specifically between the genres in the document collection and the tasks of the searcher, with an aim to exploit such relationships to improve workplace information retrieval. Research was conducted through multiple studies in three phases, moving from an exploratory study of workplace information behaviour to a controlled experimental user study. Findings confirm that workplace context shapes search behaviour. This relationship is modeled as a set of key contextual factors and sets of context-dependent access constraints, preferred document characteristics, and search strategies. Among the contextual factors identified, work tasks and information tasks were found to be significantly associated with document genres. This task-genre relationship was modeled as a matrix of associations between domain-specific task and genre taxonomies and successfully implemented as a filtering component in a workplace search system. This is the first major study of the relationship between task and genre in information seeking and of its application to information retrieval systems.