Trading spaces : on the lore and limitations of latent semantic analysis

Two decades after its inception, Latent Semantic Analysis(LSA) has become part and parcel of every modern introduction to Information Retrieval. For any tool that matures so quickly, it is important to check its lore and limitations, or else stagnation will set in. We focus here on the three main aspects of LSA that are well accepted, and the gist of which can be summarized as follows: (1) that LSA recovers latent semantic factors underlying the document space, (2) that such can be accomplished through lossy compression of the document space by eliminating lexical noise, and (3) that the latter can best be achieved by Singular Value Decomposition. For each aspect we performed experiments analogous to those reported in the LSA literature and compared the evidence brought to bear in each case. On the negative side, we show that the above claims about LSA are much more limited than commonly believed. Even a simple example may show that LSA does not recover the optimal semantic factors as intended in the pedagogical example used in many LSA publications. Additionally, and remarkably deviating from LSA lore, LSA does not scale up well: the larger the document space, the more unlikely that LSA recovers an optimal set of semantic factors. On the positive side, we describe new algorithms to replace LSA (and more recent alternatives as pLSA, LDA, and kernel methods) by trading its l2 space for an l1 space, thereby guaranteeing an optimal set of semantic factors. These algorithms seem to salvage the spirit of LSA as we think it was initially conceived.

An effective approach to verbose queries using a limited dependencies language model

Intuitively, any ‘bag of words’ approach in IR should benefit from taking term dependencies into account. Unfortunately, for years the results of exploiting such dependencies have been mixed or inconclusive. To improve the situation, this paper shows how the natural language properties of the target documents can be used to transform and enrich the term dependencies to more useful statistics. This is done in three steps. The term co-occurrence statistics of queries and documents are each represented by a Markov chain. The paper proves that such a chain is ergodic, and therefore its asymptotic behavior is unique, stationary, and independent of the initial state. Next, the stationary distribution is taken to model queries and documents, rather than their initial distributions. Finally, ranking is achieved following the customary language modeling paradigm. The main contribution of this paper is to argue why the asymptotic behavior of the document model is a better representation then just the document’s initial distribution. A secondary contribution is to investigate the practical application of this representation in case the queries become increasingly verbose. In the experiments (based on Lemur’s search engine substrate) the default query model was replaced by the stable distribution of the query. Just modeling the query this way already resulted in significant improvements over a standard language model baseline. The results were on a par or better than more sophisticated algorithms that use fine-tuned parameters or extensive training. Moreover, the more verbose the query, the more effective the approach seems to become.

An ontology-based mining approach for user search intent discovery

Discovering proper search intents is a vi- tal process to return desired results. It is constantly a hot research topic regarding information retrieval in recent years. Existing methods are mainly limited by utilizing context-based mining, query expansion, and user profiling techniques, which are still suffering from the issue of ambiguity in search queries. In this pa- per, we introduce a novel ontology-based approach in terms of a world knowledge base in order to construct personalized ontologies for identifying adequate con- cept levels for matching user search intents. An iter- ative mining algorithm is designed for evaluating po- tential intents level by level until meeting the best re- sult. The propose-to-attempt approach is evaluated in a large volume RCV1 data set, and experimental results indicate a distinct improvement on top precision after compared with baseline models.

Leveraging the network information for evaluating answer quality in a collaborative question answering portal

Collaborative question answering (cQA) portals such as Yahoo! Answers allow users as askers or answer authors to communicate, and exchange information through the asking and answering of questions in the network. In their current set-up, answers to a question are arranged in chronological order. For effective information retrieval, it will be advantageous to have the users’ answers ranked according to their quality. This paper proposes a novel approach of evaluating and ranking the users’answers and recommending the top-n quality answers to information seekers. The proposed approach is based on a user-reputation method which assigns a score to an answer reflecting its answer author’s reputation level in the network. The proposed approach is evaluated on a dataset collected from a live cQA, namely, Yahoo! Answers. To compare the results obtained by the non-content-based user-reputation method, experiments were also conducted with several content-based methods that assign a score to an answer reflecting its content quality. Various combinations of non-content and content-based scores were also used in comparing results. Empirical analysis shows that the proposed method is able to rank the users’ answers and recommend the top-n answers with good accuracy. Results of the proposed method outperform the content-based methods, various combinations, and the results obtained by the popular link analysis method, HITS.

Automata modeling for cognitive interference in users' relevance judgement

Quantum theory has recently been employed to further advance the theory of information retrieval (IR). A challenging research topic is to investigate the so called quantum-like interference in users’ relevance judgement process, where users are involved to judge the relevance degree of each document with respect to a given query. In this process, users’ relevance judgement for the current document is often interfered by the judgement for previous documents, due to the interference on users’ cognitive status. Research from cognitive science has demonstrated some initial evidence of quantum-like cognitive interference in human decision making, which underpins the user’s relevance judgement process. This motivates us to model such cognitive interference in the relevance judgement process, which in our belief will lead to a better modeling and explanation of user behaviors in relevance judgement process for IR and eventually lead to more user-centric IR models. In this paper, we propose to use probabilistic automaton(PA) and quantum finite automaton (QFA), which are suitable to represent the transition of user judgement states, to dynamically model the cognitive interference when the user is judging a list of documents.

Effect of topic domain and task type on Web image searching

Many user studies in Web information searching have found the significant effect of task types on search strategies. However, little attention was given to Web image searching strategies, especially the query reformulation activity despite that this is a crucial part in Web image searching. In this study, we investigated the effects of topic domains and task types on user’s image searching behavior and query reformulation strategies. Some significant differences in user’s tasks specificity and initial concepts were identified among the task domains. Task types are also found to influence participant’s result reviewing behavior and query reformulation strategies.

An aspect query language model based on query decomposition and high-order contextual term associations

In information retrieval (IR) research, more and more focus has been placed on optimizing a query language model by detecting and estimating the dependencies between the query and the observed terms occurring in the selected relevance feedback documents. In this paper, we propose a novel Aspect Language Modeling framework featuring term association acquisition, document segmentation, query decomposition, and an Aspect Model (AM) for parameter optimization. Through the proposed framework, we advance the theory and practice of applying high-order and context-sensitive term relationships to IR. We first decompose a query into subsets of query terms. Then we segment the relevance feedback documents into chunks using multiple sliding windows. Finally we discover the higher order term associations, that is, the terms in these chunks with high degree of association to the subsets of the query. In this process, we adopt an approach by combining the AM with the Association Rule (AR) mining. In our approach, the AM not only considers the subsets of a query as “hidden” states and estimates their prior distributions, but also evaluates the dependencies between the subsets of a query and the observed terms extracted from the chunks of feedback documents. The AR provides a reasonable initial estimation of the high-order term associations by discovering the associated rules from the document chunks. Experimental results on various TREC collections verify the effectiveness of our approach, which significantly outperforms a baseline language model and two state-of-the-art query language models namely the Relevance Model and the Information Flow model

Automated cross-lingual link discovery in Wikipedia

At NTCIR-9, we participated in the cross-lingual link discovery (Crosslink) task. In this paper we describe our approaches to discovering Chinese, Japanese, and Korean (CJK) cross-lingual links for English documents in Wikipedia. Our experimental results show that a link mining approach that mines the existing link structure for anchor probabilities and relies on the “translation” using cross-lingual document name triangulation performs very well. The evaluation shows encouraging results for our system.

Re-engaging marginalized youth through digital music production: performance, audience and evaluation

This paper presents two case studies of marginalised youth experimenting with digital music production in flexible education settings. The cases were drawn from a three-year study of alternative assessment in flexible learning centres that enrol 650+ students who have left formal schooling in Queensland, Australia. The cases are framed in reference to the literature on cultural studies approaches to education and the digital arts. Each case describes the student’s histories, cultural background and experiences, music productions, evidence of learning and re-engagement with education. Findings document how digital music production can re-engage and extend participation among students who have left formal education. They do so by theorising the online judgements and blog comments about the digital music production as a social field of exchange. It also raises critical questions about the adequacy of current approaches to evaluating and accounting for the learning and development of such youth, especially where this has occurred through creative arts and digital production.

Post-graduate health promotion students assess their information literacy

Purpose - The purpose of this paper is to examine post-graduate health promotion students’ self-perceptions of information literacy skills prior to, and after completing PILOT, an online information literacy tutorial. Design/methodology/approach – Post graduate students at Queensland University of Technology enrolled in PUP038 New Developments in Health Promotion completed a pre- and post- self-assessment questionnaire. From 2008-2011 students were required to rate their academic writing and research skills before and after completing the PILOT online information literacy tutorial. Quantitative trends and qualitative themes were analysed to establish students’ self-assessment and the effectiveness of the PILOT tutorial. Findings – The results from four years of post-graduate students’ self-assessment questionnaires provide evidence of perceived improvements in information literacy skills after completing PILOT. Some students continued to have trouble with locating quality information and analysis as well as issues surrounding referencing and plagiarism. Feedback was generally positive and students’ responses indicated they found the tutorial highly beneficial in improving their research skills. Originality/value - This paper is original because it describes post-graduate health promotion students’ self-assessment of information literacy skills over a period of four years. The literature is limited in the health promotion domain and self-assessment of post-graduate students’ information literacy skills. Keywords – Self-assessment, Post-graduate, Information literacy, Library instruction, Higher education, Health promotion, Evidence-based practice Paper Type - Research paper

Is the unigram relevance model term independent? Classifying term dependencies in query expansion

This paper develops a framework for classifying term dependencies in query expansion with respect to the role terms play in structural linguistic associations. The framework is used to classify and compare the query expansion terms produced by the unigram and positional relevance models. As the unigram relevance model does not explicitly model term dependencies in its estimation process it is often thought to ignore dependencies that exist between words in natural language. The framework presented in this paper is underpinned by two types of linguistic association, namely syntagmatic and paradigmatic associations. It was found that syntagmatic associations were a more prevalent form of linguistic association used in query expansion. Paradoxically, it was the unigram model that exhibited this association more than the positional relevance model. This surprising finding has two potential implications for information retrieval models: (1) if linguistic associations underpin query expansion, then a probabilistic term dependence assumption based on position is inadequate for capturing them; (2) the unigram relevance model captures more term dependency information than its underlying theoretical model suggests, so its normative position as a baseline that ignores term dependencies should perhaps be reviewed.

Supporting research at QUT : a tale of three librarians and a Creative Industries super-faculty

In 2012 the existing eight disciplines of Creative Industries Faculty, QUT combined with the School of Design (formerly a component of the Faculty of Built Environment and Engineering) to create a super faculty that includes the following disciplines: Architecture, Creative Writing & Literary Studies, Dance, Drama, Fashion, Film & Television, Industrial Design, Interior Design, Journalism, Media & Communication, Landscape Architecture, Music & Sound and Urban Design. The university’s research training unit AIRS (Advanced Information Retrieval Skills) is a systematic introduction to research level information literacies. It is currently being redesigned to reflect today’s new data intensive research environment and facilitate the capacity for life-long learning. Upon completion participants are expected to be able to: 1. Demonstrate an understanding of the theory of advanced search and evaluative strategies to efficiently yield appropriate resources to create original research. 2. Apply appropriate data management strategies to organise and utilize your information proficiently, ethically and legally. 3. Identify strategies to ensure best practice in the use of information sources, information technologies, information access tools and investigative methods. All Creative Industries Faculty research students must complete this unit into which CI Librarians teach discipline specific material. The library employs a team of research specific experts as well as Liaison Librarians for each faculty. Together they develop and deliver a generic research training program that provides researcher training in the following areas: Managing Research Data, QUT ePrints: New features for tracking your research impact, Tracking Research Impact, Research Students and the Library: Overview of Library Research Support Services, Technologies for Research Collaboration, Open Access Publishing, Greater Impact via Creative Commons Licence, CAMBIA - Navigating the patent literature, Uploading Publications to QUT ePrints Workshop, AIRS for supervisors, Finding Existing Research Data, Keeping up to date:Discovering and managing current awareness information and Getting Published. In 2011 Creative Industries initiated a new faculty specific research training program to promote capacity building for research within their Faculty, with workshops designed and developed with Faculty Research Leaders, The Office of Research and Liaison Librarians. “Show me the money” which assists staff to pursue alternative funding sources was one such session that was well attended and generated much discussion and interest. Drop in support sessions for ePrints, EndNote referencing software and Tracking Research Impact for the Creative Industries were also popular options on the menu. Liaison Librarians continue to provide one-on-one consultations with individual researchers as requested. This service assists Librarians greatly with getting to know and monitoring their researchers’ changing needs. The CI Faculty has enlisted two Research Leaders, one for each of the two Schools (Design and Media, Entertainment & Creative Arts) whose role it is to mentor newer research staff. Similarly within the CI library liaison team one librarian is assigned the role of Research Coordinator, whose responsibility it is to be the primary liaison with the Assistant Dean, Research and other key Faculty research managers and is the one most likely to attend Faculty committees and meetings relating to research support.

Many paths lead to discovery : analogical retrieval of cancer therapies

This paper addresses the issue of analogical inference, and its potential role as the mediator of new therapeutic discoveries, by using disjunction operators based on quantum connectives to combine many potential reasoning pathways into a single search expression. In it, we extend our previous work in which we developed an approach to analogical retrieval using the Predication-based Semantic Indexing (PSI) model, which encodes both concepts and the relationships between them in high-dimensional vector space. As in our previous work, we leverage the ability of PSI to infer predicate pathways connecting two example concepts, in this case comprising of known therapeutic relationships. For example, given that drug x TREATS disease z, we might infer the predicate pathway drug x INTERACTS WITH gene y ASSOCIATED WITH disease z, and use this pathway to search for drugs related to another disease in similar ways. As biological systems tend to be characterized by networks of relationships, we evaluate the ability of quantum-inspired operators to mediate inference and retrieval across multiple relations, by testing the ability of different approaches to recover known therapeutic relationships. In addition, we introduce a novel complex vector based implementation of PSI, based on Plate’s Circular Holographic Reduced Representations, which we utilize for all experiments in addition to the binary vector based approach we have applied in our previous research.

Semantic judgement of medical concepts : combining syntagmatic and paradigmatic information with the tensor encoding model

This paper outlines a novel approach for modelling semantic relationships within medical documents. Medical terminologies contain a rich source of semantic information critical to a number of techniques in medical informatics, including medical information retrieval. Recent research suggests that corpus-driven approaches are effective at automatically capturing semantic similarities between medical concepts, thus making them an attractive option for accessing semantic information. Most previous corpus-driven methods only considered syntagmatic associations. In this paper, we adapt a recent approach that explicitly models both syntagmatic and paradigmatic associations. We show that the implicit similarity between certain medical concepts can only be modelled using paradigmatic associations. In addition, the inclusion of both types of associations overcomes the sensitivity to the training corpus experienced by previous approaches, making our method both more effective and more robust. This finding may have implications for researchers in the area of medical information retrieval.