Adaptive Information Retrieval Research Articles

Developing and Validating a Naturalistic Decision Model for Intelligent Language-Based Decision Aids

People make mundane and critical consumption decisions every day using choice processes that are inherently constructive in nature, where preferences emerge ‘on the spot’ or ‘on the go’ using multiple strategies based on the task at hand (Bettman, Luce, & Payne, 1998; Sproule & Archer, 2000). This implies that applying a single, invariant algorithm will not solve decision problems that humans face (Tversky, Sattath, & Slovic, 1988). Instead, consumers need adaptive, multi-strategy decision aids since they shift between multiple strategies in a single decision as they acquire increasing information during the decision-making process (Bettman et al., 1998). This paper puts forth a cognitive computing approach to develop and validate a naturalistic decision model for designing language-based, mobile decision-aids (MoDA©) based on adaptive and intelligent information retrieval and multi-decision strategy use. The approach integrates established psychological theories, Elaboration Likelihood Model (ELM) and Construal Level Theory (CLT), to develop the scientific base for predicting decision-making under contingencies. ELM delineates whether human information processing is effortful or heuristic based on a person’s ability and motivation to engage in an object-relevant elaboration (Petty & Cacioppo, 1981). CLT determines whether the cognitive construal of the decision object is abstract or concrete based on psychological distance (Liberman, Trope, & Wakslak, 2007). Integrating the derivatives of these theories, the Human-Elaboration-Object-Construal (H-E-O-C) Contingency Decision Model’s central thesis is that the decision-making strategy employed by a decision-maker can be predicted by using natural language cues to infer the extent of human elaboration (low-high) on the decision and the type of knowledge (abstract-concrete) possessed on the decision object. Specifically, an extensive (vs. limited) decision strategy is likely to be employed when human elaboration revealed through natural language cues is high (vs. low). Further, an attribute-based (vs. alternative-based) strategy may be employed when the cognitive representation of the decision object is abstract (vs. concrete). Based on this theorizing, the H-E-O-C Contingency Decision Model can predict the use of four common decision strategies that systematically differ based on the amount (extensive vs. limited) and pattern (attribute- vs. alternative-based) of processing: Lexicographic or LEX (limited, attribute-based processing), Satisficing or SAT (limited, alternative-based processing), Elimination-by-Aspects or EBA (extensive, attribute-based processing), and Weighted Adding or WADD (extensive, alternative-based processing) (Bettman et al., 1998). To validate the H-E-O-C Contingency Decision Model, we conducted observational studies that simulated in-store purchase decision-making with real consumers. A total of 48 shopping sessions (n = 48) were held in a simulation home improvement retail store, and decision-making dialog between consumers and a customer service agent (trained research assistant) was recorded using wearable voice recorders. To ensure that there were fairly equal numbers of consumers who were either motivated or not to elaborate on their decisions, we created two shopping conditions – low risk (replacement AC filter purchase) and high risk (AC filter purchase to address allergy and asthma). The recorded decision dialogs were first transcribed verbatim, resulting 48 units of analysis, which were then analyzed using the grounded theory approach through open and axial coding processes (Corbin & Strauss, 1990). The open coding first identified the construal level, which was followed by axial coding to infer the decision strategy (LEX, EBA, SAT, or WADD) employed by the consumer at the initial and final stages of decision-making. This process was conducted by two coders with adequate inter-coder reliability. Two different coders coded the transcripts for the elaboration level (low vs. high) of the consumer based on specific definitions, with adequate inter-coder reliability. The H-E-O-C Contingency Decision Model proposes that high elaboration consumers will employ either WADD or EBA, whereas low elaboration consumers will employ either SAT or LEX. This proposition was supported in over 80% of the decision transcripts, offering an important validation of the framework. The main contribution of the H-E-O-C Contingency Decision Model is that it is derived from universal psychological constructs and predicts decision-making strategies that apply to many types of products and services related to healthcare, education, and finance that are characterized by attributes and alternatives. This ensures its broad applicability across a wide variety of disciplines and use cases.

Predicting users' domain knowledge in information retrieval using multiple regression analysis of search behaviors

User domain knowledge affects search behaviors and search success. Predicting a user's knowledge level from implicit evidence such as search behaviors could allow an adaptive information retrieval system to better personalize its interaction with users. This study examines whether user domain knowledge can be predicted from search behaviors by applying a regression modeling analysis method. We identify behavioral features that contribute most to a successful prediction model. A user experiment was conducted with 40 participants searching on task topics in the domain of genomics. Participant domain knowledge level was assessed based on the users' familiarity with and expertise in the search topics and their knowledge of MeSH (Medical Subject Headings) terms in the categories that corresponded to the search topics. Users' search behaviors were captured by logging software, which includes querying behaviors, document selection behaviors, and general task interaction behaviors. Multiple regression analysis was run on the behavioral data using different variable selection methods. Four successful predictive models were identified, each involving a slightly different set of behavioral variables. The models were compared for the best on model fit, significance of the model, and contributions of individual predictors in each model. Each model was validated using the split sampling method. The final model highlights three behavioral variables as domain knowledge level predictors: the number of documents saved, the average query length, and the average ranking position of the documents opened. The results are discussed, study limitations are addressed, and future research directions are suggested.

Dynamic Linking and Personalization on the Web using Linked Open Data

This paper presents a novel Semantic Web browser, called SemWeB. SemWeB supports Web browsing using Linked Open Data and personalization. This is the first time a Semantic Web browser combines the advances in Adaptive Hypermedia (i.e. personalization) and Linked Open Data. In particular, users are provided with a personalized and semantically rich Web browsing experience. For example, SemWeB supplies goal-based adaptive information retrieval from the LOD, adaptive link recommendation using a novel semantic relatedness measure and adaptive content creation. A user based study was used to assess the value of LOD-based hyperlinks and personally relevant content compared with a standard Web browser (Firefox). Results showed that users were able to browse and view more relevant information, as well as, value significance of LOD-based hyperlinks and personalized content in comparison to a standard Web browser.

Adaptive information retrieval system via modelling user behaviour

There has been an exponential growth in the volume and variety of information available on the Internet, similarly there has been a significant demand from users’ for accurate information that matches their interests, however, the two are often incompatible because of the effectiveness of retrieving the exact information the user requires. This paper addresses this problem with an adaptive agent-based modelling approach that relies on evolutionary user-modelling. The proposed information retrieval system learns user needs from user-provided relevance feedback. It is proposed that retrieval effectiveness can be improved by applying computational intelligence techniques for modelling information needs, through interactive reinforcement learning. The method combines qualitative (subjective) user relevance feedback with quantitative (algorithmic) measures of the relevance of retrieved documents. An adaptive information retrieval system is developed whose retrieval effectiveness is evaluated using traditional precision and recall.

About Our Authors

About Our Authors

User relevance criteria choices and the information search process

Relevance judgments occur within an information search process, where time, context and situation can impact the judgments. The determination of relevance is dependent on a number of factors and variables which include the criteria used to determine relevance. The relevance judgment process and the criteria used to make those judgments are manifestations of the cognitive changes which occur during the information search process. Understanding why these relevance criteria choices are made, and how they vary over the information search process can provide important information about the dynamic relevance judgment process. This information can be used to guide the development of more adaptive information retrieval systems which respond to the cognitive changes of users during the information search process. The research data analyzed here was collected in two separate studies which examined a subject’s relevance judgment over an information search process. Statistical analysis was used to examine these results and determine if there were relationships between criteria selections, relevance judgments, and the subject’s progression through the information search process. Findings confirm and extend findings of previous studies, providing strong statistical evidence of an association between the information search process and the choices of relevance criteria by users, and identifying specific changes in the user preferences for specific criteria over the course of the information search process.

The evaluation of adaptive and personalised information retrieval systems: a review

A current problem with the research of adaptive systems is the inconsistency of evaluation applied to the adaptive systems. However, evaluating an adaptive system is a difficult task due to the complexity of such systems. Evaluators need to ensure correct evaluation methods and measurement metrics are used. This paper reviews a variety of evaluation techniques applied in adaptive and user-adaptive systems. More specifically, it focuses on the user-centred evaluation of adaptive systems such as personalised recommender systems and adaptive information retrieval systems. The review tackles the question of i¾˜How have user-centred evaluations of adaptive and user-adaptive systems been conducted and how can these evaluation practices be improved?' Based on the analysed results of the: (a) evaluation approaches, (b) user-centred evaluation techniques, and (c) evaluation metrics, we propose an evaluation framework for end-user experience in evaluating adaptive systems (EFEx).

Minimizing time risk in on-line bidding: An adaptive information retrieval based approach

Knowledge is of prime importance, particularly for the individuals who are involved in e-business. A lot of energy and time is wasted by the individuals in seeking required knowledge and information. In order to facilitate the individuals with required information, an efficient technique for the proper retrieval of knowledge is must. Almost all online business activities, particularly e-auction based firms are surrounded by various risk factors pertaining to time, security, brand etc. The main focus of the present paper is to analyze all such risk factors and further to categorize the same as per their degree of influence. A nominal group technique (NGT) based approach has been utilized to do the same that ranks the risk factors using agreed criteria based approach. Further, the paper proposed an adaptive information retrieval to resolve the problems related to time risk in online bidding process, while other risk factors has been tried to resolved by using corporate memory based data warehousing. Efficient knowledge retrieval along with the knowledge development and knowledge management became a backbreaking task for any organization. A corporate memory based approach has been utilized to represent the required knowledge stored in memory warehouse for its current and future usage. In underlying retrieval model, adaptiveness is achieved using genetic algorithm based matching function adaptation, where, a total of five matching functions viz. Jaccard’s coefficient, Overlap’s coefficient, Dice coefficient, Inclusion measure, and Cosine measures have been considered to determine the retrieval effectiveness. Later, effectiveness of information retrieval system is calculated in terms of well known parameters namely precision, recall, fallout and miss. Results of adaptive information retrieval using a weighted combination of matching functions are compared with individual matching functions.

Usagers et recherche d'information. Modélisation des usagers pour la recherche d'information adaptative

User profiles are the main component of adaptive information retrieval systems. We are identifying in this paper main limits of the usage of user profiles. We are discussing about these limits in underlying some solutions and taking into account two points of view: the application point of view and the user point of view.

AIR 2008

This report describes AIR 2008, the second international workshop on Adaptive Information Retrieval (AIR), held in October 2008 at London, UK. The workshop attracted over 30 participants across the world and the largest workshop held in conjunction with IIiX 2008. The presentation ranged widely from theories to practices on Adaptive Information Retrieval. The workshop consisted of two keynote presentations, three oral presentations, breakout sessions, and poster session.

An adaptive fuzzy information retrieval model to improve response time perceived by e-commerce clients

In this paper, an adaptive fuzzy logic-based information retrieval model is presented to enable users retrieve exact and specific information they sort after. The proposed IR model takes into consideration the limited bandwidth between ISP and its users; and the characteristics (processor speed, memory size, resolution, availability of anti-virus, etc.) of clients’ devices in ensuring that a customer has a fruitful and eventful session while conducting business online. The model was designed using unified modelling language and implemented using Borland JBuilder. A performance evaluation of the proposed information retrieval system using two evaluation measures was conducted. The experimental result indicated that the model has an acceptable performance.

AIR 2006

Adaptive Information Retrieval (IR) systems are designed to optimize retrieval effectiveness and user interaction in an underlying search environment. In this article, we report on a workshop held at Glasgow, UK, in October 2006. The workshop program consisted of three invited talks, four provocative presentations, and a panel discussion. In addition, 15 poster presentations enriched the workshop themes. Important pointers to further development of adaptive IR systems and their evaluation were suggested.

Adapting information retrieval systems to user queries

Users enter queries that are short as well as long. The aim of this work is to evaluate techniques that can enable information retrieval (IR) systems to automatically adapt to perform better on such queries. By adaptation we refer to (1) modifications to the queries via user interaction, and (2) detecting that the original query is not a good candidate for modification. We show that the former has the potential to improve mean average precision (MAP) of long and short queries by 40% and 30% respectively, and that simple user interaction can help towards this goal. We observed that after inspecting the options presented to them, users frequently did not select any. We present techniques in this paper to determine beforehand the utility of user interaction to avoid this waste of time and effort. We show that our techniques can provide IR systems with the ability to detect and avoid interaction for unpromising queries without a significant drop in overall performance.

Towards a belief-revision-based adaptive and context-sensitive information retrieval system

In an adaptive information retrieval (IR) setting, the information seekers' beliefs about which terms are relevant or nonrelevant will naturally fluctuate. This article investigates how the theory of belief revision can be used to model adaptive IR. More specifically, belief revision logic provides a rich representation scheme to formalize retrieval contexts so as to disambiguate vague user queries. In addition, belief revision theory underpins the development of an effective mechanism to revise user profiles in accordance with information seekers' changing information needs. It is argued that information retrieval contexts can be extracted by means of the information-flow text mining method so as to realize a highly autonomous adaptive IR system. The extra bonus of a belief-based IR model is that its retrieval behavior is more predictable and explanatory. Our initial experiments show that the belief-based adaptive IR system is as effective as a classical adaptive IR system. To our best knowledge, this is the first successful implementation and evaluation of a logic-based adaptive IR model which can efficiently process large IR collections.

Information Processing & Management Special Topic Issue on Adaptive Information Retrieval

Information Processing & Management Special Topic Issue on Adaptive Information Retrieval

“Conversational” information systems

PurposeThe purpose of this paper is to review recent developments in educational informatics relating to the provision by information systems of pedagogical support to web‐based learners, and to propose further investigation of the feasibility and potential value of web‐based “conversational” information systems to complement adaptive hypermedia and information retrieval systems.Design/methodology/approachThe potential of Pask's conversation theory is considered as a potentially useful framework for the development of information systems capable of providing pedagogical support for web‐based learners, complementary to that provided by existing computer‐assisted learning and adaptive hypermedia systems. The potential role and application of entailment meshes are reviewed in relation to other forms of knowledge representation including classifications, semantic networks, ontologies and representations based on knowledge space theory.FindingsConcludes that conversation theory could be a useful framework to support the development of web‐based “conversational” information that would complement aspects of computer‐assisted learning, adaptive hypermedia and information retrieval systems. The entailment mesh knowledge representation associated with conversation theory provides the potential for providing particularly rich pedagogical support by virtue of its properties of cyclicity, consistency and connectivity, designed to support deep and enduring levels of understanding.Research limitations/implicationsAlthough based on a considerable body of theoretical and empirical work relating to conversation theory, the paper remains speculative in that the gap is still great between, on the one hand, what has so far been achieved and, on the other, the practical realisation of its potential to enhance web‐based learning. Much work remains to be done in terms of exploring the extent to which procedures developed and benefits found in relatively small‐scale experimental contexts can effectively be scaled to yield enhanced support for “real world” learning‐related information behaviour.Originality/valueThe ideas of Pask, discussed in this paper, are capable of guiding the structuring of information according to parameters designed to facilitate deep and enduring understanding via interactive “conversational” engagement between the conceptual structures of information source authors and learners. If one can scale Pask's work to “real world” learning‐related information behaviour, one can increase the range of web‐based information systems and services capable of providing pedagogical support to web‐based learners.

Maxi-Adjustment and Possibilistic Deduction for Adaptive Information Agents

The expressive power of logic is believed to be able to model most of the fundamental aspects of information retrieval (IR). However, it is also understood that classical logic is ineffective for handling partiality and uncertainty in IR. Applying non-classical logics such as the AGM belief revision logic and the possibilistic logic to adaptive information retrieval is appealing since they provide a powerful and rigorous framework to model partiality and uncertainty inherent in any IR processes. The maxi-adjustment method, which is an effective computational apparatus of the AGM paradigm, is applied to develop the learning components of the adaptive information agents. Essentially, maxi-adjustment allows the partial representation K of a user's information needs N to be refined gradually based on the user's relevance feedback t. Generally speaking, learning in adaptive information agents is characterised by the AGM belief revision Kt*. On the other hand, possibilistic logic supports a gradated assessment of the uncertainty arising from matching K with the imperfect characterisation d of an information object D. Information matching in adaptive information agents is underpinned by K ƕ d, where ƕ is the possibilistic inference relation. This paper illustrates how maxi-adjustment and possibilistic deduction can be applied to develop the adaptive information agents. Their impact on the agents learning autonomy and explanatory power is also discussed.

An information retrieval system based on a user profile

In this era of information explosion, providing the right information to the right person within reasonable time duration is a very important goal for today’s information retrieval (IR) systems. Due to the characteristics of retrieving methods, the conventional IR models often suffer from inaccurate and incomplete queries as well as inconsistent document relevance. Moreover, each user has his/her own interpretation of the semantic meaning of query terms during the retrieving process. Thus, high accuracy of retrieved information is somewhat hard to achieve. However, accuracy can be improved by designing an IR model that can adapt to the diverse needs of an individual and can perform a personalized search. In this paper, an adaptive information retrieval system embedded in an intelligent feedback tuning mechanism is proposed to capture the personal notions of query terms. This mechanism together with a correlation table is used as a user profile to ‘simulate’ the user’s notions of terms. This correlation table includes the degrees of semantic relevance (SR) and co-occurrence (CO) among terms. By using the personal profile during retrieval, the system can obtain a better match between the information needs and the retrieved results. It is also shown that the recall/precision rates for our system are approximately 91% and 81%, respectively, on average.

A fuzzy genetic algorithm approach to an adaptive information retrieval agent

We present an approach to a Genetic Information Retrieval Agent Filter (GIRAF) for documents from the Internet using a genetic algorithm (GA) with fuzzy set genes to learn the user’s information needs. The population of chromosomes with fixed length represents such user’s preferences. Each chromosome is associated with a fitness that may be considered the system’s belief in the hypothesis that the chromosome, as a query, represents the user’s information needs. In a chromosome, every gene characterizes documents by a keyword and an associated occurrence frequency, represented by a certain type of a fuzzy subset of the set of positive integers. Based on the user’s evaluation of the documents retrieved by the chromosome, compared to the scores computed by the system, the fitness of the chromosomes is adjusted. A prototype of GIRAF has been developed and tested. The results of the test are discussed, and some directions for further works are pointed out.

A fuzzy genetic algorithm approach to an adaptive information retrieval agent

We present an approach to a Genetic Information Retrieval Agent Filter (GIRAF) for documents from the Internet using a genetic algorithm (GA) with fuzzy set genes to learn the user's information needs. The population of chromosomes with fixed length represents such user's preferences. Each chromosome is associated with a fitness that may be considered the system's belief in the hypothesis that the chromosome, as a query, represents the user's information needs. In a chromosome, every gene characterizes documents by a keyword and an associated occurrence frequency, represented by a certain type of a fuzzy subset of the set of positive integers. Based on the user's evaluation of the documents retrieved by the chromosome, compared to the scores computed by the system, the fitness of the chromosomes is adjusted. A prototype of GIRAF has been developed and tested. The results of the test are discussed, and some directions for further works are pointed out.