A data mining framework for relevance feature discovery

This thesis is a study for automatic discovery of text features for describing user information needs. It presents an innovative data-mining approach that discovers useful knowledge from both relevance and non-relevance feedback information. The proposed approach can largely reduce noises in discovered patterns and significantly improve the performance of text mining systems. This study provides a promising method for the study of Data Mining and Web Intelligence.

The heterogeneous cluster ensemble method using hubness for clustering text documents

We propose a cluster ensemble method to map the corpus documents into the semantic space embedded in Wikipedia and group them using multiple types of feature space. A heterogeneous cluster ensemble is constructed with multiple types of relations i.e. document-term, document-concept and document-category. A final clustering solution is obtained by exploiting associations between document pairs and hubness of the documents. Empirical analysis with various real data sets reveals that the proposed meth-od outperforms state-of-the-art text clustering approaches.

Reliability and validity of query intent assessments

In most intent recognition studies, annotations of query intent are created post hoc by external assessors who are not the searchers themselves. It is important for the field to get a better understanding of the quality of this process as an approximation for determining the searcher's actual intent. Some studies have investigated the reliability of the query intent annotation process by measuring the interassessor agreement. However, these studies did not measure the validity of the judgments, that is, to what extent the annotations match the searcher's actual intent. In this study, we asked both the searchers themselves and external assessors to classify queries using the same intent classification scheme. We show that of the seven dimensions in our intent classification scheme, four can reliably be used for query annotation. Of these four, only the annotations on the topic and spatial sensitivity dimension are valid when compared with the searcher's annotations. The difference between the interassessor agreement and the assessor-searcher agreement was significant on all dimensions, showing that the agreement between external assessors is not a good estimator of the validity of the intent classifications. Therefore, we encourage the research community to consider using query intent classifications by the searchers themselves as test data.

Acquiring search expertise : learning experiences and threshold concepts

Expert searchers engage with information as information brokers, researchers, reference librarians, information architects, faculty who teach advanced search, and in a variety of other information-intensive professions. Their experiences are characterized by a profound understanding of information concepts and skills and they have an agile ability to apply this knowledge to interacting with and having an impact on the information environment. This study explored the learning experiences of searchers to understand the acquisition of search expertise. The research question was: What can be learned about becoming an expert searcher from the learning experiences of proficient novice searchers and highly experienced searchers? The key objectives were: (1) to explore the existence of threshold concepts in search expertise; (2) to improve our understanding of how search expertise is acquired and how novice searchers, intent on becoming experts, can learn to search in more expertlike ways. The participant sample drew from two population groups: (1) highly experienced searchers with a minimum of 20 years of relevant professional experience, including LIS faculty who teach advanced search, information brokers, and search engine developers (11 subjects); and (2) MLIS students who had completed coursework in information retrieval and online searching and demonstrated exceptional ability (9 subjects). Using these two groups allowed a nuanced understanding of the experience of learning to search in expertlike ways, with data from those who search at a very high level as well as those who may be actively developing expertise. The study used semi-structured interviews, search tasks with think-aloud narratives, and talk-after protocols. Searches were screen-captured with simultaneous audio-recording of the think-aloud narrative. Data were coded and analyzed using NVivo9 and manually. Grounded theory allowed categories and themes to emerge from the data. Categories represented conceptual knowledge and attributes of expert searchers. In accord with grounded theory method, once theoretical saturation was achieved, during the final stage of analysis the data were viewed through lenses of existing theoretical frameworks. For this study, threshold concept theory (Meyer & Land, 2003) was used to explore which concepts might be threshold concepts. Threshold concepts have been used to explore transformative learning portals in subjects ranging from economics to mathematics. A threshold concept has five defining characteristics: transformative (causing a shift in perception), irreversible (unlikely to be forgotten), integrative (unifying separate concepts), troublesome (initially counter-intuitive), and may be bounded. Themes that emerged provided evidence of four concepts which had the characteristics of threshold concepts. These were: information environment: the total information environment is perceived and understood; information structures: content, index structures, and retrieval algorithms are understood; information vocabularies: fluency in search behaviors related to language, including natural language, controlled vocabulary, and finesse using proximity, truncation, and other language-based tools. The fourth threshold concept was concept fusion, the integration of the other three threshold concepts and further defined by three properties: visioning (anticipating next moves), being light on one's 'search feet' (dancing property), and profound ontological shift (identity as searcher). In addition to the threshold concepts, findings were reported that were not concept-based, including praxes and traits of expert searchers. A model of search expertise is proposed with the four threshold concepts at its core that also integrates the traits and praxes elicited from the study, attributes which are likewise long recognized in LIS research as present in professional searchers. The research provides a deeper understanding of the transformative learning experiences involved in the acquisition of search expertise. It adds to our understanding of search expertise in the context of today's information environment and has implications for teaching advanced search, for research more broadly within library and information science, and for methodologies used to explore threshold concepts.

Towards modelling web search behaviour : integrating users’ cognitive styles

With the rapid growth of information on the Web, the study of information searching has let to an increased interest. Information behaviour (IB) researchers and information systems (IS) developers are continuously exploring user - Web search interactions to understand and to help users to provide assistance with their information searching. In attempting to develop models of IB, several studies have identified various factors that govern user's information searching and information retrieval (IR), such as age, gender, prior knowledge and task complexity. However, how users' contextual factors, such as cognitive styles, affect Web search interactions has not been clearly explained by the current models of Web Searching and IR. This study explores the influence of users' cognitive styles on their Web search behaviour. The main goal of the study is to enhance Web search models with a better understanding of how these cognitive styles affect Web searching. Modelling Web search behaviour with a greater understanding of user's cognitive styles can help information science researchers and IS designers to bridge the semantic gap between the user and the IS. To achieve the aims of the study, a user study with 50 participants was conducted. The study adopted a mixed method approach incorporating several data collection strategies to gather a range of qualitative and quantitative data. The study utilised pre-search and post-search questionnaires to collect the participants' demographic information and their level of satisfaction about the search interactions. Riding's (1991) Cognitive Style Analysis (CSA) test was used to assess the participants' cognitive styles. Participants completed three predesigned search tasks and the whole user - web search interactions, including thinkaloud, were captured using a monitoring program. Data analysis involved several qualitative and quantitative techniques: the quantitative data gave raise to detailed findings about users' Web searching and cognitive styles, the qualitative data enriched the findings with illustrative examples. The study results provide valuable insights into Web searching behaviour among different cognitive style users. The findings of the study extend our understanding of Web search behaviour and how users search information on the Web. Three key study findings emerged: • Users' Web search behaviour was demonstrated through information searching strategies, Web navigation styles, query reformulation behaviour and information processing approaches while performing Web searches. The manner in which these Web search patterns were demonstrated varied among the users with different cognitive style groups. • Users' cognitive styles influenced their information searching strategies, query reformulation behaviour, Web navigational styles and information processing approaches. Users with particular cognitive styles followed certain Web search patterns. • Fundamental relationships were evident between users' cognitive styles and their Web search behaviours; and these relationships can be illustrated through modelling Web search behaviour. Two models that depict the associations between Web search interactions, user characteristics and users' cognitive styles were developed. These models provide a greater understanding of Web search behaviour from the user perspective, particularly how users' cognitive styles influence their Web search behaviour. The significance of this research is twofold: it will provide insights for information science researchers, information system designers, academics, educators, trainers and librarians who want to better understand how users with different cognitive styles perform information searching on the Web; at the same time, it will provide assistance and support to the users. The major outcomes of this study are 1) a comprehensive analysis of how users search the Web; 2) extensive discussion on the implications of the models developed in this study for future work; and 3) a theoretical framework to bridge high-level search models and cognitive models.

XML clustering and its application to XML transformation

The continuous growth of the XML data poses a great concern in the area of XML data management. The need for processing large amounts of XML data brings complications to many applications, such as information retrieval, data integration and many others. One way of simplifying this problem is to break the massive amount of data into smaller groups by application of clustering techniques. However, XML clustering is an intricate task that may involve the processing of both the structure and the content of XML data in order to identify similar XML data. This research presents four clustering methods, two methods utilizing the structure of XML documents and the other two utilizing both the structure and the content. The two structural clustering methods have different data models. One is based on a path model and other is based on a tree model. These methods employ rigid similarity measures which aim to identifying corresponding elements between documents with different or similar underlying structure. The two clustering methods that utilize both the structural and content information vary in terms of how the structure and content similarity are combined. One clustering method calculates the document similarity by using a linear weighting combination strategy of structure and content similarities. The content similarity in this clustering method is based on a semantic kernel. The other method calculates the distance between documents by a non-linear combination of the structure and content of XML documents using a semantic kernel. Empirical analysis shows that the structure-only clustering method based on the tree model is more scalable than the structure-only clustering method based on the path model as the tree similarity measure for the tree model does not need to visit the parents of an element many times. Experimental results also show that the clustering methods perform better with the inclusion of the content information on most test document collections. To further the research, the structural clustering method based on tree model is extended and employed in XML transformation. The results from the experiments show that the proposed transformation process is faster than the traditional transformation system that translates and converts the source XML documents sequentially. Also, the schema matching process of XML transformation produces a better matching result in a shorter time.

Modeling users’ web search behavior and their cognitive styles

Previous studies have shown that users’ cognitive styles play an important role during Web searching. However, only limited studies have showed the relationship between cognitive styles and Web search behavior. Most importantly, it is not clear which components of Web search behavior are influenced by cognitive styles. This paper examines the relationships between users’ cognitive styles and their Web searching and develops a model that portrays the relationship. The study uses qualitative and quantitative analyses to inform the study results based on data gathered from 50 participants. A questionnaire was utilised to collect participants’ demographic information, and Riding’s (1991) Cognitive Style Analysis (CSA) test to assess their cognitive styles. Results show that users’ cognitive styles influenced their information searching strategies, query reformulation behaviour, Web navigational styles and information processing approaches. The user model developed in this study depicts the fundamental relationships between users’ Web search behavior and their cognitive styles. Modeling Web search behavior with a greater understanding of user’s cognitive styles can help information science researchers and information systems designers to bridge the semantic gap between the user and the systems. Implications of the research for theory and practice, and future work are discussed.

Rapid yet accurate speech indexing using dynamic match lattice spotting

The support for typically out-of-vocabulary query terms such as names, acronyms, and foreign words is an important requirement of many speech indexing applications. However, to date many unrestricted vocabulary indexing systems have struggled to provide a balance between good detection rate and fast query speeds. This paper presents a fast and accurate unrestricted vocabulary speech indexing technique named Dynamic Match Lattice Spotting (DMLS). The proposed method augments the conventional lattice spotting technique with dynamic sequence matching, together with a number of other novel algorithmic enhancements, to obtain a system that is capable of searching hours of speech in seconds while maintaining excellent detection performance

A topic based document relevance ranking model

Topic modelling has been widely used in the fields of information retrieval, text mining, machine learning, etc. In this paper, we propose a novel model, Pattern Enhanced Topic Model (PETM), which makes improvements to topic modelling by semantically representing topics with discriminative patterns, and also makes innovative contributions to information filtering by utilising the proposed PETM to determine document relevance based on topics distribution and maximum matched patterns proposed in this paper. Extensive experiments are conducted to evaluate the effectiveness of PETM by using the TREC data collection Reuters Corpus Volume 1. The results show that the proposed model significantly outperforms both state-of-the-art term-based models and pattern-based models.

Detecting news topics from microblogs using sequential pattern mining

This thesis presents a sequential pattern based model (PMM) to detect news topics from a popular microblogging platform, Twitter. PMM captures key topics and measures their importance using pattern properties and Twitter characteristics. This study shows that PMM outperforms traditional term-based models, and can potentially be implemented as a decision support system. The research contributes to news detection and addresses the challenging issue of extracting information from short and noisy text.

Consensus σ70 promoter prediction using hadoop

MapReduce frameworks such as Hadoop are well suited to handling large sets of data which can be processed separately and independently, with canonical applications in information retrieval and sales record analysis. Rapid advances in sequencing technology have ensured an explosion in the availability of genomic data, with a consequent rise in the importance of large scale comparative genomics, often involving operations and data relationships which deviate from the classical Map Reduce structure. This work examines the application of Hadoop to patterns of this nature, using as our focus a wellestablished workflow for identifying promoters - binding sites for regulatory proteins - Across multiple gene regions and organisms, coupled with the unifying step of assembling these results into a consensus sequence. Our approach demonstrates the utility of Hadoop for problems of this nature, showing how the tyranny of the "dominant decomposition" can be at least partially overcome. It also demonstrates how load balance and the granularity of parallelism can be optimized by pre-processing that splits and reorganizes input files, allowing a wide range of related problems to be brought under the same computational umbrella.

Optimization of an integrated model for automatic reduction and expansion of long queries

A long query provides more useful hints for searching relevant documents, but it is likely to introduce noise which affects retrieval performance. In order to smooth such adverse effect, it is important to reduce noisy terms, introduce and boost additional relevant terms. This paper presents a comprehensive framework, called Aspect Hidden Markov Model (AHMM), which integrates query reduction and expansion, for retrieval with long queries. It optimizes the probability distribution of query terms by utilizing intra-query term dependencies as well as the relationships between query terms and words observed in relevance feedback documents. Empirical evaluation on three large-scale TREC collections demonstrates that our approach, which is automatic, achieves salient improvements over various strong baselines, and also reaches a comparable performance to a state of the art method based on user’s interactive query term reduction and expansion.

Using the quantum probability ranking principle to rank interdependent documents

A known limitation of the Probability Ranking Principle (PRP) is that it does not cater for dependence between documents. Recently, the Quantum Probability Ranking Principle (QPRP) has been proposed, which implicitly captures dependencies between documents through “quantum interference”. This paper explores whether this new ranking principle leads to improved performance for subtopic retrieval, where novelty and diversity is required. In a thorough empirical investigation, models based on the PRP, as well as other recently proposed ranking strategies for subtopic retrieval (i.e. Maximal Marginal Relevance (MMR) and Portfolio Theory(PT)), are compared against the QPRP. On the given task, it is shown that the QPRP outperforms these other ranking strategies. And unlike MMR and PT, one of the main advantages of the QPRP is that no parameter estimation/tuning is required; making the QPRP both simple and effective. This research demonstrates that the application of quantum theory to problems within information retrieval can lead to significant improvements.

The interactive PRP for diversifying document rankings

The assumptions underlying the Probability Ranking Principle (PRP) have led to a number of alternative approaches that cater or compensate for the PRP's limitations. In this poster we focus on the Interactive PRP (iPRP), which rejects the assumption of independence between documents made by the PRP. Although the theoretical framework of the iPRP is appealing, no instantiation has been proposed and investigated. In this poster, we propose a possible instantiation of the principle, performing the first empirical comparison of the iPRP against the PRP. For document diversification, our results show that the iPRP is significantly better than the PRP, and comparable to or better than other methods such as Modern Portfolio Theory.

Back to the roots : mean-variance analysis of relevance estimations

Recently, mean-variance analysis has been proposed as a novel paradigm to model document ranking in Information Retrieval. The main merit of this approach is that it diversifies the ranking of retrieved documents. In its original formulation, the strategy considers both the mean of relevance estimates of retrieved documents and their variance. How- ever, when this strategy has been empirically instantiated, the concepts of mean and variance are discarded in favour of a point-wise estimation of relevance (to replace the mean) and of a parameter to be tuned or, alternatively, a quantity dependent upon the document length (to replace the variance). In this paper we revisit this ranking strategy by going back to its roots: mean and variance. For each retrieved document, we infer a relevance distribution from a series of point-wise relevance estimations provided by a number of different systems. This is used to compute the mean and the variance of document relevance estimates. On the TREC Clueweb collection, we show that this approach improves the retrieval performances. This development could lead to new strategies to address the fusion of relevance estimates provided by different systems.