Abstract

Finding similar patents is a challenging task in patent information retrieval. A patent application is often a starting point to find similar inventions. Keyword search for similar patents requires significant domain expertise and may not fetch relevant results. We propose a novel representation for patents and use a two stage approach to find similar patents. Each patent is represented as an IPC class vector. Citation network of patents is used to propagate these vectors from a node (patent) to its neighbors (cited patents). Thus, each patent is represented as a weighted combination of its IPC information as well as of its neighbors. A query patent is represented as a vector using its IPC information and similar patents can be simply found by comparing this vector with vectors of patents in the corpus. Text based search is used to re-rank this solution set to improve precision. We experiment with two similarity measures and re-ranking strategies to empirically show that our representation is effective in improving both precision and recall of queries of CLEF-2011 dataset.

Abstract

In web search, understanding the user intent plays an important role in improving search experience of the end users. Such an intent can be represented by the categories which the user query belongs to. In this work, we propose an information retrieval based approach to query categorization with an emphasis on learning category rankings. To carry out categorization we first represent a category by web documents (from Open Directory Project) that describe the semantics of the category. Then, we learn the category rankings for the queries using 'learning to rank' techniques. To show that the results obtained are consistent and do not vary across datasets, we evaluate our approach on two datasets including the publicly available KDD Cup dataset. We report an overall improvement of 20% on all evaluation metrics (precision, recall and F-measure) over two baselines: a text categorization baseline and an unsupervised IR baseline.

Abstract

The world of E-commerce is expanding, posing a large arena of products, their descriptions, customer and professional reviews that are pertinent to them. Most of the product attribute extraction techniques in literature work on structured descriptions using several text analysis tools. However, attributes in these descriptions are limited compared to those in customer reviews of a product, where users discuss deeper and more specific attributes. In this paper, we propose a novel supervised domain independent model for product attribute extraction from user reviews. The user generated content contains unstructured and semi-structured text where conventional language grammar dependent tools like parts-of-speech taggers, named entity recognizers, parsers do not perform at expected levels. We used Wikipedia and Web to identify product attributes from customer reviews and achieved F1score of 0.73.

Abstract

Sentiment mining and classification plays an important role in predicting what people think about products, places, etc. In this piece of work, using basic NLP Techniques like NGram, POS-Tagged NGram we classify movie and product reviews broadly into two polarities: Positive and Negative. We propose a model to address the problem of determining whether a review is positive or negative, we experiment and use several machine learning algorithms Naive Bayes (NB), Multi-Layer Perceptron (MLP), Support Vector Machine (SVM) to have a comparative study of the performance of the method we devised in this work. Along with this we also did negation handling and observed improvements in classification. The algorithm we proposed achieved an average accuracy of 78.32% on movie and 70.06% on multi-category dataset. In this paper we focus on the collective study of Ngram and POS tagged information available in the reviews.

Abstract

A synergistic approach is described combining computer simulation with experiment in the interpretation of small angle neutron scattering (SANS) and inelastic scattering experiments on the structure and dynamics of proteins and other biopolymers. Simulation models can be tested by calculating neutron scattering structure factors and comparing the results directly with experimental scattering profiles. If the scattering profiles agree the simulations can be used to provide a detailed decomposition and interpretation of the experiments, and if not, the models can be rationally adjusted. Comparison with neutron experiment can be made at the level of the scattering functions or, less directly, of structural and dynamical quantities derived from them. This methodology is discussed in the context of the protein glass transition, protein-solvent dynamical coupling, the density of the hydration shell of proteins and the structural analysis of lignocellulosic biomass.

Abstract

In this report, we present details about the participation of IIIT Hyderabad in Guided Summarization and Knowledge Base Population tracks at TAC 2011. we have enhanced our summarization system with knowledge based measures. Wikipedia based extraction methods and topic modelling are used to score sentences in guided summarization track. For multilingual summarization task, we investigated the HAL (Hyperspace Analogue to Language Model) where we created a semantic space from word co-occurrences. We show that the results obtained with this unsupervised language independent method are competitive with other state-of-the-art systems. For monolingual and multilingual entity linking task, we extended our previous year’s model to a light weight language independent system without utilizing any other external knowledge or resource.

Abstract

MapReduce framework has received a wide acclaim over the past few years for large scale computing. It has become a standard paradigm for batch oriented workloads. As the adoption of this paradigm has increased rapidly, scheduling of these MapReduce jobs has become a problem of great interest in research community. We propose an approach which tries to maintain harmony among the jobs running on the cluster, and in turn decrease their runtime. In our model, the scheduler is made aware of different types of jobs running on the cluster. The scheduler tries to allocate a task on a node if the incoming task does not affect the tasks already running on that node. From the list of available pending tasks, our algorithm selects the one that is most compatible with the tasks already running on that node. We bring up heuristic and machine learning based solutions to our approach and try to maintain a resource balance on the cluster by not overloading any of the nodes, thereby reducing the overall runtime of the jobs. The results show a saving of runtime of around 21% in the case of heuristic based approach and around 27% in the case of machine learning based approach when compared to Yahoo's Capacity scheduler.

Abstract

Frequent patterns are an important class of regularities thatexist in a transaction database. Certain frequent patternswith low minimum support (minsup) value can provide use-ful information in many real-world applications. However,extraction of these frequent patterns with singleminsup-based frequent pattern mining algorithms such as Aprioriand FP-growth leads to “rare item problem.” That is, athighminsupvalue, the frequent patterns with lowminsupare missed, and at lowminsupvalue, the number of frequentpatterns explodes. In the literature,“multipleminsupsframe-work” was proposed to discover frequent patterns. Further-more, frequent pattern mining techniques such as MultipleSupport Apriori and Conditional Frequent Pattern-growth(CFP-growth) algorithms have been proposed. As the fre-quent patterns mined with this framework do not satisfydownward closure property, the algorithms follow differenttypes of pruning techniques to reduce the search space. Inthis paper, we propose an efficient CFP-growth algorithmby proposing new pruning techniques. Experimental resultsshow that the proposed pruning techniques are effective.

Abstract

Recommender systems employ the popular K-nearest neigh-bor collaborative filtering (K-CF) methodology and its vari-ations for recommending the products. In K-CF approach,recommendation for a given user is computed based on theratings of K-nearest neighbors. In K-CF approach, it can benoted that, the system identifiesKneighbors for each userirrespective of the number of products he/she has rated.As a result, the user who have rated few products may getthe less-similar neighbors and the user who have rated moreproducts may miss the genuine neighbors. In the literature,the notion of lower-bound similarity has been proposed toimprove the clustering performance in which the clusters areextracted by fixing the similarity threshold. In this paper,we have extended the notion of lower-bound similarity torecommender systems to improve the performance of K-CFapproach. In the proposed approach, instead of fixingKfor finding the neighborhood, the similarity threshold valueis fixed to extract the neighbors for each user. As a re-sult, each user gets appropriate number of neighbors basedon the number of products rated by him/her in a dynamicmanner. The experimental results, on MovieLens dataset,show that the proposed lower bound similarity CF approachimproves the performance of recommender systems over K-CF approach.

Abstract

In this paper, we have made an effort to propose approachesto find similar legal judgements by extending the popu-lar techniques used in information retrieval and search en-gines. Legal judgements are complex in nature and re-fer other judgements. We have analyzed all-term, legal-term, co-citation and bibliographic coupling-based similaritymethods to find similar judgements. The experimental re-sults show that the legal-term cosine similarity method per-forms better than all-term cosine similarity method. Also,the results show that bibliographic coupling similarity methodimproves the performance over co-citation approach.