IIT Madras is a public technical and research university located in Chennai, Tamil Nadu. Founded in 1959, it is recognised as an Institute of National Importance.

IIT Madras has been ranked as the top engineering institute in India for four years in a row by the National Institutional Ranking Framework of the MHRD

It currently offers undergraduate, postgraduate and research degrees across 16 disciplines in Engineering, Sciences, Humanities and Management. About 596 faculty belonging to science and engineering departments and centres of the Institute are engaged in teaching, research and industrial consultancy.

IIT Madras

Conversational Recommender Systems belong to a class of knowledge based systems which simulate a customer's interaction with a shopkeeper with the help of repeated user feedback till the user settles on a product. One of the modes for getting user feedback is Preference Based Feedback, which is especially suited for novice users(having little domain knowledge), who find it easy to express preferences across products as a whole, rather than specific product features. Such kind of novice users might not be aware of the specific characteristics of the items that they may be interested in, hence, the shopkeeper/system should show them a set of products during each interaction, which can constructively stimulate their preferences, leading them to a desirable product in subsequent interactions. We propose a novel approach to conversational recommendation, Util- Sim, where utilities corresponding to products get continually updated as a user iteratively interacts with the system, helping her discover her hidden preferences in the process. We show that UtilSim, which combines domain-specific "dominance" knowledge with SimRank based similarity, significantly outperforms the existing conversational approaches using Preference Based Feedback in terms of recommendation efficiency. © Springer-Verlag Berlin Heidelberg 2013.

GENERAL

Department of Computer Science and Engineering

Lecture Notes in Business Information Processing

Utilsim: Iteratively helping users discover their preferences

In this paper, we propose Latent Dirichlet Allocation (LDA) [1] based document classification algorithm which does not require any labeled dataset. In our algorithm, we construct a topic model using LDA, assign one topic to one of the class labels, aggregate all the same class label topics into a single topic using the aggregation property of the Dirichlet distribution and then automatically assign a class label to each unlabeled document depending on its "closeness" to one of the aggregated topics. We present an extension to our algorithm based on the combination of Expectation-Maximization (EM) algorithm and a naive Bayes classifier. We show effectiveness of our algorithm on three real world datasets. Copyright © 2013 ACM.

SIGIR 2013 - Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval

Document classification by topic labeling

Conversational Recommendation mimics the kind of dialog that takes between a customer and a shopkeeper involving multiple interactions where the user can give feedback at every interaction as opposed to Single Shot Retrieval, which corresponds to a scheme where the system retrieves a set of items in response to a user query in a single interaction. Compromise refers to a particular user preference which the recommender system failed to satisfy. But in the context of conversational systems, where the user's preferences keep on evolving as she interacts with the system, what constitutes as a compromise for her also keeps on changing. Typically, in Single Shot retrieval, the notion of compromise is characterized by the assignment of a particular feature to a particular dominance group such as MIB (higher value is better) or LIB (lower value is better) and this assignment remains true for all the users who use the system. In this paper, we propose a way to realize the notion of compromise in a conversational setting. Our approach, Flexi-Comp, introduces the notion of dynamically assigning a feature to two dominance groups simultaneously which is then used to redefine the notion of compromise. We show experimentally that a utility function based on this notion of compromise outperforms the existing conversational rec-ommenders in terms of recommendation efficiency. Copyright 2013 ACM. 15.00.

International Conference on Information and Knowledge Management, Proceedings

Flexible and dynamic compromises for effective recommendations

Queries submitted to a search engine are usually short and ambiguous. Query expansion is an effective way to resolve the ambiguity of a query. We attempt to make a consolidated study of the effectiveness of query logs in this regard. We report empirical findings on the effect of personalization over an anonymous query log. We have designed a novel evaluation method to quantify improvements due to personalization. In the second part of our work, we present a novel approach to intelligently detect queries whose expansions potentially lead to improvements in retrieval. Empirical results obtained using this approach clearly show significant improvements over indiscriminate query expansion. Copyright © 2013 ACM.

Towards selective user specific query expansion

Textual problem-solution repositories are available today in various forms, most commonly as problem-solution pairs from community question answering systems. Modern search engines that operate on the web can suggest possible completions in real-time for users as they type in queries. We study the problem of generating intelligent query suggestions for users of customized search systems that enable querying over problem-solution repositories. Due to the small scale and specialized nature of such systems, we often do not have the luxury of depending on query logs for finding query suggestions. We propose a retrieval model for generating query suggestions for search on a set of problem solution pairs. We harness the problem solution partition inherent in such repositories to improve upon traditional query suggestion mechanisms designed for systems that search over general textual corpora. We evaluate our technique over real problem-solution datasets and illustrate that our technique provides large and statistically significant improvements over the state-of-the-art technique in query suggestion. © 2013 Springer-Verlag.

Fulltext

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Query suggestions for textual problem solution repositories

When a user of a microblogging site authors a microblog post or browses through a microblog post, it provides cues as to what topic she is interested in at that point in time. Example-based search that retrieves similar tweets given one exemplary tweet, such as the one just authored, can help provide the user with relevant content. We investigate various components of microblog posts, such as the associated timestamp, author's social network, and the content of the post, and develop approaches that harness such factors in finding relevant tweets given a query tweet. An empirical analysis of such techniques on real world twitter-data is then presented to quantify the utility of the various factors in assessing tweet relevance. We observe that content-wise similar tweets that also contain extra information not already present in the query, are perceived as useful. We then develop a composite technique that combines the various approaches by scoring tweets using a dynamic query-specific linear combination of separate techniques. An empirical evaluation establishes the effectiveness of the composite technique, and that it outperforms each of its constituents. © 2012 Springer-Verlag.

Finding relevant tweets

This paper describes a novel approach for automatic creation of Bangla error corpus for training and evaluation of grammar checker systems. The procedure begins with automatic creation of large number of erroneous sentences from a set of grammatically correct sentences. A statistical Confidence Score Filter has been implemented to select proper samples from the generated erroneous sentences such that sentences with less probable word sequences get lower confidence score and vice versa. Rule based Mal-rule filter with HMM based semi-supervised POS tagger has been used to collect the sentences having improper tag sequences. Combination of these two filters ensures the robustness of the proposed approach such that no valid construction is getting selected within the synthetically generated error corpus. Though the present work focuses on the most frequent grammatical errors in Bangla written text, detail taxonomy of grammatical errors in Bangla is also presented here, with an aim to increase the coverage of the error corpus in future. The proposed approach is language independent and could be easily applied for creating similar corpora in other languages. © 2012 Springer-Verlag.

Combining confidence score and Mal-rule filters for automatic creation of Bangla error corpus: Grammar checker perspective

The quality of the cases maintained in a case base has a direct influence on the quality of the proposed solutions. The presence of cases that do not conform to the similarity hypothesis decreases the alignment of the case base and often degrades the performance of a CBR system. It is therefore important to find out the suitability of each case for the application of CBR and associate a solution with a certain degree of confidence. Feature weighting is another important aspect that determines the success of a system, as the presence of irrelevant and redundant attributes also results in incorrect solutions. We explore these problems in conjunction with a real-world CBR application called InfoChrom. It is used to predict the values of several soil nutrients based on features extracted from a chromatogram image of a soil sample. We propose novel feature weighting techniques based on alignment, as well as a new alignment and confidence measure as potential solutions. The hypotheses are evaluated on UCI datasets and the case base of Infochrom and show promising results. © 2012 Springer-Verlag.

Feature weighting and confidence based prediction for case based reasoning systems

In this paper, we look into the problem of filtering problem solution repositories (from sources such as community-driven question answering systems) to render them more suitable for usage in knowledge reuse systems. We explore harnessing the fuzzy nature of usability of a solution to a problem, for such compaction. Fuzzy usabilities lead to several challenges; notably, the trade-off between choosing generic or better solutions. We develop an approach that can heed to a user specification of the trade-off between these criteria and introduce several quality measures based on fuzzy usability estimates to ascertain the quality of a problem-solution repository for usage in a Case Based Reasoning system. We establish, through a detailed empirical analysis, that our approach outperforms state-of-the-art approaches on virtually all quality measures. © 2011 ACM.

More or better: On trade-offs in compacting textual problem solution repositories

This paper explores how background knowledge from freely available web resources can be utilised for Textual Case Based Reasoning. The work reported here extends the existing Explicit Semantic Analysis approach to representation, where textual content is represented using concepts with correspondence to Wikipedia articles. We present approaches to identify Wikipedia pages that are likely to contribute to the effectiveness of text classification tasks. We also study the effect of modelling semantic similarity between concepts (amounting to Wikipedia articles) empirically. We conclude with the observation that integrating background knowledge from resources like Wikipedia into TCBR tasks holds a lot of promise as it can improve system effectiveness even without elaborate manual knowledge engineering. Significant performance gains are obtained using a very small number of features that have very strong correspondence to how humans describe the domain. © 2011 Springer-Verlag.

Selective integration of background knowledge in TCBR systems

The performance of a Textual Case-Based Reasoning system is critically dependent on its underlying model of text similarity, which in turn is dependent on similarity between terms and phrases in the domain. In the absence of human intervention, term similarities are often modelled using co-occurrence statistics, which are fragile unless the corpus is truly representative of the domain. We present the case for introspective revision in TCBR, whereby the system incrementally revises its term similarity knowledge by exploiting conflicts of its representation against an alternate source of knowledge such as category knowledge in classification tasks, or linguistic and background knowledge. The advantage of such revision is that it requires no human intervention. Our experiments on classification knowledge show that revision can lead to substantial gains in classification accuracy, with results competitive to best-in-line text classifiers. We have also presented experimental results over synthetic data to suggest that the idea can be extended to improve case-base alignment in TCBR domains with textual problem and solution descriptions. © 2010 Springer-Verlag.

Introspective knowledge revision in textual case-based reasoning

In TCBR, complexity refers to the extent to which similar problems have similar solutions. Casebase complexity measures proposed are based on the premise that a casebase is simple if similar problems have similar solutions. We observe, however, that such measures are vulnerable to choice of solution side representations, and hence may not be meaningful unless similarities between solution components of cases are shown to corroborate with human judgements. In this paper, we redefine the goal of complexity measurements and explore issues in estimating solution side similarities. A second limitation of earlier approaches is that they critically rely on the choice of one or more parameters. We present two parameter-free complexity measures, and propose a visualization scheme for casebase maintenance. Evaluation over diverse textual casebases show their superiority over earlier measures. © 2009 Springer Berlin Heidelberg.

Robust measures of complexity in TCBR

Recently, stochastic simulations of networks of chemical reactions have shown distributions of steady states that are inconsistent with the steady state solutions of the corresponding deterministic ordinary differential equations. One such class of systems is comprised of networks that have irreversible reactions, and the origin of the anomalous behavior in these cases is understood to be due to the existence of absorbing states. More puzzling is the report of such anomalies in reaction networks without irreversible reactions. One such biologically important example is the futile cycle. Here we show that, in these systems, nonclassical behavior can originate from a stochastic elimination of all the molecules of a key species. This leads to a reduction in the topology of the network and the sampling of steady states corresponding to a truncated network. Surprisingly, we find that, in spite of the purely discrete character of the topology reduction mechanism revealed by "exact" numerical solutions of the master equations, this phenomenon is reproduced by the corresponding Fokker-Planck equations. © 2009 American Institute of Physics.

Department of Aerospace Engineering

World University root node

Journal of Chemical Physics

Stochastic bimodalities in deterministically monostable reversible chemical networks due to network topology reduction

Textual-case based reasoning (TCBR) systems where the problem and solution are in free text form are hard to evaluate. In the absence of class information, domain experts are needed to evaluate solution quality, and provide relevance information. This approach is costly and time consuming. We propose three measures that can be used to compare alternate TCBR system configurations, in the absence of class information. The main idea is to quantify alignment as the degree to which similar problems have similar solutions. Two local measures capture this information by analysing similarity between problem and solution neighbourhoods at different levels of granularity, whilst a global measure achieves the same by analyzing similarity between problem and solution clusters. We determine the suitability of the proposed measures by studying their correlation with classifier accuracy on a health and safety incident reporting task. Strong correlation is observed with all three approaches with local measures being slightly superior over the global one. © Springer-Verlag Berlin Heidelberg 2008.

Evaluation measures for TCBR systems

This paper deals with two relatively less well studied problems in Textual CBR, namely visualizing and evaluating complexity of textual case bases. The first is useful in case base maintenance, the second in making informed choices regarding case base representation and tuning of parameters for the TCBR system, and also for explaining the behaviour of different retrieval/classification techniques over diverse case bases. We present an approach to visualize textual case bases by "stacking" similar cases and features close to each other in an image derived from the case-feature matrix. We propose a complexity measure called GAME that exploits regularities in stacked images to evaluate the alignment between problem and solution components of cases. GAME class , a counterpart of GAME in classification domains, shows a strong correspondence with accuracies reported by standard classifiers over classification tasks of varying complexity. © Springer-Verlag Berlin Heidelberg 2008.

Visualizing and evaluating complexity of textual case bases

Latent Semantic Indexing (LSI) has been shown to be effective in recovering from synonymy and polysemy in text retrieval applications. However, since LSI ignores class labels of training documents, LSI generated representations are not as effective in classification tasks. To address this limitation, a process called 'sprinkling' is presented. Sprinkling is a simple extension of LSI based on augmenting the set of features using additional terms that encode class knowledge. However, a limitation of sprinkling is that it treats all classes (and classifiers) in the same way. To overcome this, we propose a more principled extension called Adaptive Sprinkling (AS). AS leverages confusion matrices to emphasise the differences between those classes which are hard to separate. The method is tested on diverse classification tasks, including those where classes share ordinal or hierarchical relationships. These experiments reveal that AS can significantly enhance the performance of instance-based techniques (kNN) to make them competitive with the state-of-the-art SVM classifier. The revised representations generated by AS also have a favourable impact on SVM performance.

IJCAI International Joint Conference on Artificial Intelligence

Supervised latent semantic indexing using adaptive sprinkling

Latent Semantic Indexing (LSI) is an established dimensionality reduction technique for Information Retrieval applications. However, LSI generated dimensions are not optimal in a classification setting, since LSI fails to exploit class labels of training documents. We propose an approach that uses class information to influence LSI dimensions whereby class labels of training documents are endoded as new terms, which are appended to the documents. When LSI is carried out on the augmented term-document matrix, terms pertaining to the same class are pulled closer to each other. Evaluation over experimental data reveals significant improvement in classification accuracy over LSI. The results also compare favourably with naive Support Vector Machines. © Springer-Verlag Berlin Heidelberg 2006.

Sprinkling: Supervised latent semantic indexing

Case Retrieval Networks (CRNs) facilitate flexible and efficient retrieval in Case-Based Reasoning (CBR) systems. While CRNs scale up well to handle large numbers of cases in the case-base, the retrieval efficiency is still critically determined by the number of feature values (referred to as Information Entities) and by the nature of similarity relations defined over the feature space. In textual domains it is typical to perform retrieval over large vocabularies with many similarity interconnections between words. This can have adverse effects on retrieval efficiency for CRNs. This paper proposes an extension to CRN, called the Fast Case Retrieval Network (FCRN) that eliminates redundant computations at run time. Using artificial and real-world datasets, it is demonstrated that FCRNs can achieve significant retrieval speedups over CRNs, while maintaining retrieval effectiveness. © Springer-Verlag Berlin Heidelberg 2006.

Fast case retrieval nets for textual data

Problem solving with experiences that are recorded in text form requires a mapping from text to structured cases, so that case comparison can provide informed feedback for reasoning. One of the challenges is to acquire an indexing vocabulary to describe cases. We explore the use of machine learning and statistical techniques to automate aspects of this acquisition task. A propositional semantic indexing tool, Psi, which forms its indexing vocabulary from new features extracted as logical combinations of existing keywords, is presented. We propose that such logical combinations correspond more closely to natural concepts and are more transparent than linear combinations. Experiments show Psi-derived case representations to have superior retrieval performance to the original keyword-based representations. Psi also has comparable performance to Latent Semantic Indexing, a popular dimensionality reduction technique for text, which unlike Psi generates linear combinations of the original features.

European Conference on Principles of Data Mining and Knowledge Discovery

A Propositional Approach to Textual Case Indexing

Intelligent directory assistance is the ability to retrieve relevant subscriber records in the presence of mismatches between the query and the subscriber record. The challenges are to provide inexact retrieval over very large data volumes within a reasonable timeframe as well as the ability to tailor retrieval to differing query requirements. In this paper we discuss a case-based approach to intelligent directory assistance which focusses on a layered approach to retrieval. The case-base is multiply indexed where each level indexes the case-base at a level of dissimilarity. We introduce the concepts of Levels of Lenience, Degree of Lenience and Search templates and demonstrate how these enable retrieval to be fast and yet tailored to query requirements. A production quality system based on these ideas has been implemented at a leading telecommunications service provider and is giving excellent performance.