Abstract

In this paper we report our performance at DUC 2007 summarization tasks. We participated both in the query-focused multidocument summarization main task and in a pilot update summary generation tasks. This year we used a term clustering approach to better estimate a sentence prior. We used only the sentence prior which is query independent, in the update summarization task and found that it’s performance is comparable with the top performing systems. In the main task our system ranked 1 in ROUGE-2, ROUGE-SU4 and ROUGE-BE scores as well as in pyramid scores.

Abstract

Commonly used vocabulary in Indian language documents found on the web contain a number of words that have Sanskrit, Persian or English origin. However, such words may be written in different scripts with slight variations in spelling and morphology. In this paper we explore approximate string matching techniques to exploit this situation of relatively large number of cognates among Indian languages, which are higher when compared to an Indian language and a non-Indian language. We present an approach to identify cognates and make use of them for improving dictionary based CLIR when the query and documents both belong to two different Indian languages. We conduct experiments using a Hindi document collection and a set of Telugu queries and report the improvement due to cognate recognition and translation.

Abstract

Named Entity Recognition (NER) is the task of identifying and classifying all proper nouns in a document as person names, or—ganization names, location names, date & time expressions and miscellaneous. Previ—ous work (Cucerzan and Yarowsky, 1999) was done using the complete words as fea—tures which suffers from a low recall prob—lem. Character n—gram based approach (Klein et al., 2003) using generative mod—els, was experimented on English language and it proved to be useful over the word based models. Applying the same technique on Indian Languages, we experimented with Conditional Random Fields (CRFs), a dis—criminative model, and evaluated our sys—tem on two Indian Languages Telugu and Hindi. The character n—gram based models showed considerable improvement over the word based models. This paper describes the features used and experiments to increase the recall of Named Entity Recognition Systems which is also language independent

Abstract

A cross language query based multi-document summarization task requires a system to produce a brief, coherent and readable summary serving the information need represented in the query. In this paper we present a Cross Language Query Based Multi Document Summarization system for Telugu to English language pair. The topics being in Telugu and the documents from which summary is to be generated are in English. The system uses a Telugu to English bilingual lexicon to obtain translations for each word as well as uses some transliteration techniques and a target language relevance based language modeling using Hyper Space Analogue to Language. The system was evaluated using DUC 2005 topics translated into Telugu and the English model summaries provided at DUC 2005 and the results are discussed in the evaluation section.

Abstract

In recent times, information technology (IT) has a tremendous influence on how different Indian government departments operate. At one level it has given rise to a requirement for timely delivery of information and services to the citizen round the clock. At a higher level, it enables the citizen to conduct transactions through government portals. This changed the purpose of the government portals from primarily a place to advertise the success stories of government to an effective means to provide efficient services to the citizen. As the needs of the citizen shift and IT capabilities evolve, government departments are under increasing pressure to become more innovative - both in terms of government processes and their delivery model. One of the important factors to provide efficient government services is availability of all government services through a single point delivery platform. The government departments should have an integrated platform with a secured access control system. It is thus important to design a coherent enterprise architecture framework for the government departments. The aim of this research paper is to outline an effective enterprise architecture framework and an innovative technological solution that can serve as the common platform for provision of all government services to the citizen of India.

Abstract

Many knowledge workers are increasingly using online resources to find out latest developments in their specialty and articles of interest. To extract relevant information from such multiple online information sources summarization is being used. Current summarization systems produce a uniform version of summary for all users. However summaries which are generic in nature do not cater to the userpsilas background and interests. In this paper we propose to make the summarization process user specific and present a design for generating personalized summaries of online articles that are tailored to each personpsilas interest. The userpsilas data available on Web is used for model their background and interest. A controlled user-centered qualitative evaluation carried out on news articles of science and technology domain, indicates better user satisfaction with personalized summaries compared to generic summaries.

Abstract

A read-only transaction (ROT) does not modify any data. The main issues regarding processing of read-only transactions (ROTs) are correctness, data currency and performance. Even though the popular two-phase locking protocol processes ROTs correctly with no data currency related issues, its performance deteriorates with data contention. To improve the performance of ROTs, snapshot isolation-based approaches have been proposed. Eventhough snapshot isolation-based approaches improve the performance of ROTs, both data currency of ROTs andcorrectness (serializability) are compromised. In this paper, we propose an asynchronous speculative locking protocol (called as ASLR) which improves the performance ofROTs by trading extra processing resources. The simulation results show that ASLR improves the performance of ROTs significantly over two-phase locking and snapshot isolation-based approaches with manageable extra processing resources. The ASLR approach processes ROTs without any data currency and correctness issues.

Abstract

A read-only transaction (ROT) does not modify anydata. Efforts are being made in the literature toimprove the performance of ROTs without correct-ness and data currency issues. The widely used two-phase locking protocol (2PL) processes the transac-tions without any correctness and data currency issues.However, the performance of 2PL deteriorates withdata contention. Snapshot isolation (SI)-based pro-tocols proposed in the literature improve the perfor-mance of ROTs, but they compromise on correctnessand data currency issues. Speculative locking (SL)protocols are proposed in the literature for improv-ing the performance of ROTs by carrying out spec-ulative executions only for ROTs and following 2PLfor update transactions. In SL-based protocols, up-date transactions are blocked if they conflicting withROTs. In this paper, we have proposed an improvedapproach to improve parallelism among update trans-actions and ROTs by exploiting a new notion called“compensatability”. In this protocol, an ROT whichcan be “compensatable” can complete the executionand carry out compensating operation to incorporatethe effect of conflicting update transactions. As aresult, the parallelism is improved over SL protocolsas the update transactions which are conflicting with‘compensatable’ ROTs need not block. In this paper,we have proposed a protocol by exploiting both “com-pensatability” property of ROTs and speculation. Thesimulation results show that the proposed protocol im-proves the performance by carrying out less number ofspeculative executions. Further, the proposed protocoldoes not violate serializability criteria.

Abstract

Prokaryotic Genomes undergo various mutational events during the course of genome evolution through a variety of processes including Horizontal Gene Transfer. These regions of about 10 - 100 Kb in length are called Genomic Islands and presence of such regions offers selective advantage to the organism and may aid in the organism's growth, survival (saprophytic or symbiotic mode) or display of pathogenicity. Earlier analysis on horizontally transferred regions has shown significant variations from the average genomic values in their GC content, codon preferences and dinucleotide bias. We have developed a web-based tool identifying genomic islands in prokaryotic genomes (GIPro) based on various measures, viz., GC content (total and at individual codon positions), dinculeotide biases, and biases in codon and amino acid usage, carried out in a sliding window analysis along the length of the genome. The regions that deviate by n standard deviations (n - user defined, default value 2) from the average genome values for at least two measures are identified as probable genomic islands by the tool. GIPro can also be used for comparison of user defined regions / gene sets to confirm whether both belong to the same genome or have differences in their dinucleotide or codon usage biases as a result of horizontal transfer.

Abstract

Identifying tandem repeats in the proteome of any organism is important not only for understanding the structure and function of the proteins but also for analyzing the association of abnormal expansion of repeat regions with disorders We have developed an efficient tool for identifying Peptide Periodic Repeats (PEPPER) in protein sequences. The tool has three in-built modules: In the first module, initial screening of the protein sequence by a uni-dimensional vector equivalent to the dot-matrix is used to identify all possible repeat regions. In the second module, a sliding window analysis of the putative repeat region using a K-tuple identifies the period and the possible pattern of the repeat. In the third module, wraparound dynamic programming (WDP) algorithm is implemented to obtain the exact repeat boundaries. If the protein sequence does not contain tandem repeat region, then the second and third steps of the algorithm are skipped making it computationally efficient for database search compared to the other existing algorithms. We report up to three possible multiple-periodicities in a given repeat region. The tool reports the consensus repeat pattern, the complete repeat region, the score and alignment of the consensus with the repeat region along with percentage mismatch/indels. A preliminary comparison of our results with the TRIPS database of protein tandem repeats showed that our program not only identifies all the repeats in the TRIPS database, but we were also able to re-annotate a few incorrectly annotated repeats and found few new repeats no reported in the TRIPS database. PEPPER provides the user the facility to change parameter values for identifying small or approximate tandem repeats. A database of repeats can be built by the user for his set of protein sequences on the fly with PEPPER.