(CSE) Department, IIT Bombay Annual Report for 2018-2019 Introduction

Total Page:16

File Type:pdf, Size:1020Kb

(CSE) Department, IIT Bombay Annual Report for 2018-2019 Introduction Computer Science and Engineering (CSE) Department, IIT Bombay Annual Report for 2018-2019 Introduction: The Computer Science and Engineering (CSE) department at the Indian Institute of Technology (IIT) Bombay is the largest among CSE Departments in any institute in India. The research and teaching in the department spans a wide spectrum of areas including algorithms, animation, artificial intelligence, compilers, combinatorial optimization, computer vision, data mining, embedded systems, formal methods, functional programming, e-commerce, graphics, databases, image processing and analysis, machine learning, medical image computing, mobile computing, natural language processing, object oriented systems, parallel and distributed processing, programming languages, reinforcement learning, query processing and optimization, real time systems, security, speech processing, software engineering, systems, theoretical computer science, wireless and sensor networks, and verification. The department has 45 faculty members, including 2 visiting faculty and 1 adjunct faculty, with Prof. Uday Khedker as the current head. Student and Faculty Research Achievements (within the last year): ● Prof. Soumen Chakrabarti receives Amazon Research Award to study "Neural Decomposition of Complex Queries Into Subqueries Over Knowledge Graph and Corpus". ● Prof. Soumen Chakrabarti has been elected as a Fellow of the Indian National Science Academy for his seminal contributions to web search and mining. ● Professor Suyash P. Awate and his students (PhD, BTech) have got 3 papers accepted at MICCAI ● Jerin Geo James Ph.D. Student guided by Prof. Ajit Rajwade has won a fellowship from Qualcomm ● Prof. Pushpak Bhattacharyya has been conferred upon the “Distinguished Alumnus Award” for 2018 by his alma mater, IIT Kharagpur. ● Two books where published by CFILT Lab Aditya Joshi, Pushpak Bhattacharyya and Mark J. Carman, Investigations in Computational Sarcasm, Cognitive Systems Monograph Series, Springer Nature, ISBN:978-981-10-8395-2, 2018. Abhijit Mishra and Pushpak Bhattacharyya, Cognitively Inspired Natural Language Processing- An Investigation Based on Eye Tracking, Cognitive Intelligence and Robotics Series, Springer Nature Singapore, ISBN:978-981-13-1515-2, 2018. ● Prof. Sundar Vishwanathan and Prof. Parag Chaudhuri from CSE have been conferred Departmental awards for Excellence in Teaching for the year 2018. ● Prof. Ashwin Gumaste has been awarded the Shanti Swarup Bhatnagar (SSB) Prize for the year 2018 in Engineering Sciences for his outstanding contributions to the development of end-to- end carrier-class networking and carrier ethernet switch routers which are deployed extensively in national infrastructure. SSB prize is a rare distinction conferred by the Government of India upon scientists who have demonstrated exceptional potential in Science and Technology. ● Four faculty members were conferred IIT Bombay Research Awards for the year 2017 announced on October 24th, 2018. ● Four faculty members were conferred IIT Bombay Research Awards for the year 2017 announced on October 24th, 2018. Prof. Supratik Chakraborty was awarded the Research Publication Award for his work on "Scalable constrained sampling and counting with strong approximation guarantees". Prof. Ganesh Ramakrishnan and Prof. Parag Chaudhuri were awarded the Impactful Research Award for their research on "Development of an adaptive framework for end-to-end corrections in Indic OCR" Prof. Siddhartha Chaudhuri was awarded the Early Research Achiever Award. ● B.Tech student Saurabh Garg guided by Prof. Suyash Awate and M.T student Sandeep Subramanian guided by Prof. Ganesh Ramakrishnan have received Excellence in Research Awards for 2018. List of Publications (Papers published in Peer-Reviewed Journals): Debajyoty Banik, Asif Ekbal, Pushpak Bhattacharyya: Machine Learning Based Optimized Pruning Approach for Decoding in Statistical Machine Translation. IEEE Access 7: 1736-1751 (2019) Naveen Saini, Sriparna Saha, Anubhav Jangra, Pushpak Bhattacharyya: Extractive single document summarization using multi-objective optimization: Exploring self-organized differential evolution, grey wolf optimizer and water cycle algorithm. Knowl.-Based Syst. 164: 45-67 (2019) Shweta Yadav, Asif Ekbal, Sriparna Saha, Ankit Kumar, Pushpak Bhattacharyya: Feature assisted stacked attentive shortest dependency path based Bi-LSTM model for protein-protein interaction. Knowl.-Based Syst. 166: 18-29 (2019) V. Rudra Murthy, Mitesh M. Khapra, Pushpak Bhattacharyya: Improving NER Tagging Performance in Low-Resource Languages via Multilingual Learning. ACM Trans. Asian & Low-Resource Lang. Inf. Process. 18(2): 9:1-9:20 (2019) Md. Shad Akhtar, Palaash Sawant, Sukanta Sen, Asif Ekbal, Pushpak Bhattacharyya: Improving Word Embedding Coverage in Less-Resourced Languages Through Multi-Linguality and Cross- Linguality: A Case Study with Aspect-Based Sentiment Analysis. ACM Trans. Asian & Low-Resource Lang. Inf. Process. 18(2): 15:1-15:22 (2019) Sabyasachi Kamila, Mohammed Hasanuzzaman, Asif Ekbal, Pushpak Bhattacharyya: Tempo-HindiWordNet: A Lexical Knowledge-base for Temporal Information Processing. ACM Trans. Asian & Low-Resource Lang. Inf. Process. 18(2): 19:1-19:22 (2019) Anoop Kunchukuttan, Mitesh M. Khapra, Gurneet Singh, Pushpak Bhattacharyya: Leveraging Orthographic Similarity for Multilingual Neural Transliteration. TACL 6: 303-316 (2018) Jonas Harbering, Abhiram Ranade, Marie Schmidt, Oliver Sinnen: Complexity, bounds and dynamic programming algorithms for single track train scheduling. Annals OR 273(1-2): 479-500 (2019) Ahana Pradhan, Rushikesh K. Joshi: A Taxonomy of Consistency Models in Dynamic Migration of Business Processes. IEEE Trans. Services Computing 11(3): 562-579 (2018) Uma Sawant, Soumen Chakrabarti, Ganesh Ramakrishnan: "Open-domain question answering using a knowledge graph and web corpus" by Uma Sawant, Soumen Chakrabarti and Ganesh Ramakrishnan with Martin Vesely as coordinator. SIGWEB Newsletter 2018(Winter): 4:1-4:8 (2018) Sridhar Iyer, Shree Prakash Singh: Multiple-period planning of Internet Protocol-over-Elastic Optical Networks. J. Information Telecommunication 3(1): 39-56 (2019) 2018 Sridhar Iyer, Shree Prakash Singh: An efficient link based model for routing, modulation format and spectrum allocation with quality of transmission consideration in elastic optical networks. IJIPT 11(3): 137-144 (2018) Sridhar Iyer, Sujata Sengar, Rochak Bajpai, Shree Prakash Singh: On the Performance of MLR Optical WDM Network Based on ITU-T Conforming Fibers in the Presence of Dominant Physical Layer Impairments. J. Comm. Inform. Networks 3(3): 91-108 (2018) Nidhi Tiwari, Umesh Bellur, Santonu Sarkar, Maria Indrawan: Optimizing MapReduce for energy efficiency. Softw., Pract. Exper. 48(9): 1660-1687 (2018) S. Akshay, Paul Gastin, Shankara Narayanan Krishna: Analyzing Timed Systems Using Tree Automata. Logical Methods in Computer Science 14(2) (2018) Dipak L. Chaudhari, Om P. Damani: Assumption propagation through annotated programs. Formal Asp. Comput. 29(3): 495-530 (2017) Debadatta Mishra, Purushottam Kulkarni: A survey of memory management techniques in virtualized systems. Computer Science Review 29: 56-73 (2018) Uma Sawant, Soumen Chakrabarti, Ganesh Ramakrishnan: "Open-domain question answering using a knowledge graph and web corpus" by Uma Sawant, Soumen Chakrabarti and Ganesh Ramakrishnan with Martin Vesely as coordinator. SIGWEB Newsletter 2018(Winter): 4:1-4:8 (2018) Suryajith Chillara, Nutan Limaye, Srikanth Srinivasan: Small-Depth Multilinear Formula Lower Bounds for Iterated Matrix Multiplication with Applications. SIAM J. Comput. 48(1): 70-92 (2019) S. Akshay, Paul Gastin, Shankara Narayanan Krishna: Analyzing Timed Systems Using Tree Automata. Logical Methods in Computer Science 14(2) (2018) Mukulika Maity, Bhaskaran Raman, Mythili Vutukuru, Avinash Chaurasia, Rachit Srivastava: Witals: AP-Centric Health Diagnosis of WiFi Networks. IEEE Trans. Mob. Comput. 17(4): 975-989 (2018) Haibin Huang, Evangelos Kalogerakis, Siddhartha Chaudhuri, Duygu Ceylan, Vladimir G. Kim, Ersin Yumer: Learning Local Shape Descriptors from Part Correspondences with Multiview Convolutional Networks. ACM Trans. Graph. 37(1): 6:1-6:14 (2018) Chenyang Zhu, Kai Xu, Siddhartha Chaudhuri, Renjiao Yi, Hao Zhang: SCORES: shape composition with recursive substructure priors. ACM Trans. Graph. 37(6): 211:1-211:14 (2018) Anshu S. Anand, R. K. Shyamasundar, Sathya Peri: STMs in practice: Partial rollback vs pure abort mechanisms. Concurrency and Computation: Practice and Experience 31(5) (2019) Stephen A. Fenner, Rohit Gurjar, Thomas Thierauf: A deterministic parallel algorithm for bipartite perfect matching. Commun. ACM 62(3): 109-115 (2019) List of Publications (Papers Published in Conferences) Ajit A. Diwan, Bodhayan Roy, Subir Kumar Ghosh: Drawing Bipartite Graphs in Two Layers with Specified Crossings. CALDAM 2019: 97-108 Prasanna Kumar K., Amitabha Sanyal, Amey Karkare, Saswat Padhi: A static slicing method for functional programs and its incremental version. CC 2019: 53-64 Prasanna Kumar K., Amitabha Sanyal, Amey Karkare, Saswat Padhi: A static slicing method for functional programs and its incremental version. CC 2019: 53-64 Shantanu Thakoor, Simoni Shah, Ganesh Ramakrishnan, Amitabha Sanyal: Synthesis of Programs from Multimodal Datasets. AAAI 2018: 184-191 Shrawan Kumar, Amitabha Sanyal, R. Venkatesh, Punit Shah: Property Checking Array Programs Using Loop Shrinking. TACAS (1) 2018: 213-231 Abhijeet Dubey, Aditya
Recommended publications
  • PRASHANT KUMAR SHARMA 173059007 Computer Science & Engineering M.Tech
    PRASHANT KUMAR SHARMA 173059007 Computer Science & Engineering M.Tech. Indian Institute of Technology Bombay Male Specialization: Computer Science and Engineering DOB: 24/12/1993 Examination University Institute Year CPI / % Post Graduation IIT Bombay IIT Bombay 2020 7.53 Undergraduate Specialization : Computer Science and Engineering Dr. A.P.J. Abdul Kalam Technical Kamla Nehru Institute of Technology, Graduation University Sultanpur 2016 84.02 Intermediate/+2 CBSE Kendriya Vidyalaya AFS Gorakhpur 2011 83.40 Matriculation CBSE Kendriya Vidyalaya AFS Gorakhpur 2009 90.80 FIELDS OF INTEREST • Machine Learning, Deep Learning, Natural Language Processing, Computer Vision, Reinforcement Learning M.TECH THESIS AND SEMINAR • Explainable Artificial Intelligence (M.Tech Project) Guide: Prof. Pushpak Bhattacharyya | Industry Collaborator: IBM (May’19 - Present) ◦ Product Objective : To develop an explainability framework which takes a trained deep learning model as input and interprets it with state-of-the-art methods like Integrated Gradients and LIME algorithm. This framework will be distributed as a python package, which can be used either as a command-line tool or as a web service. ◦ Research Objective : To develop a model to capture causal relationships in natural languages, and differentiate between causal and correlated factors using knowledge graphs and deep learning techniques. • Explainable Artificial Intelligence (M.Tech Seminar) Guide: Prof. Pushpak Bhattacharyya | Industry Collaborator: IBM (Sprint’19) ◦ Objective: : Conducted extensive research literature survey of explainability of deep neural networks: Need, Scope, Challenges and state-of-the-art methods. ◦ Involved comprehensive study of explainability of deep neural network representation, output and algorithms for explainability: Sensitivity Analysis, Layer-wise Relevance Propagation, LIME, Integrated Gradients, Influence Functions, Sanity Checks for Saliency maps, Attention mechanism, Semantically Equivalent Rules.
    [Show full text]
  • IITK Chronicle IITK Newsletter || Volume 2 Issue 2, April 2018 Highlights from the Director’S Desk
    Indian Institute of Technology Kanpur IITK Chronicle IITK Newsletter || Volume 2 Issue 2, April 2018 Highlights From the Director’s Desk e are once again at the end of another academic session and in the last two months, we have Prof. Abhay Karandikar, Department of successfully organized some of the most awaited annual events of our campus. Techkriti, our Electrical Engineering, IIT Bombay annual technological fest, saw record participation and the Institute Flower Show had the entire appointed as the Director, IIT Kanpur. W campus coming together for a fun show and fete. The institute's Women's Cell also organized great initiatives in the past month for gender sensitization. With Prof. Monika Katiyar appointed Head of these events, IIT Kanpur strives to set an example for a zero-tolerance policy towards sexual harassment on the Department of Materials Science & campus. Several of our faculty members have won notable awards and recognitions and the SIDBI Engineering. Innovation and Incubation Centre won the Platinum Award under the "Smart Incubator of the Year" category by India Smart Grid Forum. Prof. Amalendu Chandra appointed Head of the Department of Chemistry. For the coming months, we have many exciting things in the pipeline – the start of the new academic session and the Convocation always create a special buzz on campus. But most of all, we look forward to welcoming our new Director, Prof. Abhay Karandikar, and working with him to take IIT Kanpur to greater heights. New Colleagues Prof. Manindra Agrawal Officiating Director, Prof. Subrahmanyam Saderla IIT Kanpur Aerospace Engineering Prof. Ankush Sharma In Focus Electrical Engineering Techkriti 2018 Prof.
    [Show full text]
  • Word Sense Disambiguation Using Indowordnet
    Word Sense Disambiguation Using IndoWordNet Sudha Bhingardive Department of Computer Science and Engineering, Indian Institute of Technology-Bombay, Powai, Mumbai, India. Email: [email protected] Pushpak Bhattacharyya Department of Computer Science and Engineering, Indian Institute of Technology-Bombay, Powai, Mumbai, India. Email: [email protected] Abstract Word Sense Disambiguation (WSD) is considered as one of the toughest problem in the field of Natural Language Processing. IndoWordNet is a linked structure of WordNets of major Indian languages. Recently, several IndoWordNet based WSD approaches have been proposed and implemented for Indian languages. In this chapter, we present the usage of various other features of IndoWordNet in performing WSD. Here, we have used features like linked WordNets, semantic and lexical relations, etc. We have followed two unsupervised approaches, viz., (1) use of IndoWordNet in bilingual WSD for finding the sense distribution with the help of Expectation Maximization algorithm, (2) use of IndoWordNet in WSD for finding the most frequent sense using word and sense embeddings. Both these approaches justifies the importance of IndoWordNet for word sense disambiguation for Indian languages, as the results are found to be promising and can beat the baselines. Keywords: IndoWordNet, WordNet, Word Sense Disambiguation, WSD, Bilingual WSD, Unsupervised WSD, Most Frequent Sense, MFS 1. Introduction 1.1. What is Word Sense Disambiguation? Word Sense Disambiguation or WSD is the task of identifying the correct meaning of a word in a given context. The necessary condition for a word to be disambiguated is that it should have multiple senses. Generally, in order to disambiguate a given word, we should have a context in which the word has been used and knowledge about the word, otherwise it becomes difficult to get the exact meaning of a word.
    [Show full text]
  • ANNUAL REPORT 2019-20 IIT Bombay Annual Report 2019-20 Content
    IIT BOMBAY ANNUAL REPORT 2019-20 IIT BOMBAY ANNUAL REPORT 2019-20 Content 1) Director’s Report 05 2) Academic Programmes 07 3) Research and Development Activities 09 4) Outreach Programmes 26 5) Faculty Achievements and Recognitions 27 6) Student Activities 31 7) Placement 55 8) Society For Innovation And Entrepreneurship 69 9) IIT Bombay Research Park Foundation 71 10) International Relations 73 11) Alumni And Corporate Relations 84 12) Institute Events 90 13) Facilities 99 a) Infrastructure Development b) Central Library c) Computer Centre d) Centre For Distance Engineering Education Programme 14) Departments/ Centres/ Schools and Interdisciplinary Groups 107 15) Publications 140 16) Organization 141 17) Summary of Accounts 152 Director's Report By Prof. Subhasis Chaudhuri, Director, IIT Bombay Indian Institute of Technology Bombay acknowledged for their research contributions. (IIT Bombay) has a rich tradition of pursuing We have also been able to further our links with excellence and has continually re-invented international and national peer universities, itself in terms of academic programmes and enabling us to enhance research and educational research infrastructure. Students are exposed programmes at the Institute. to challenging, research-based academics and IIT Bombay continues to make forays into a host of sport, cultural and organizational newer territories pertinent to undergraduate activities on its vibrant campus. The presence and postgraduate education. At postgraduate of world-class research facilities, vigorous level, a specially designed MA+PhD dual institute-industry collaborations, international degree programme in Philosophy under the exchange programmes, interdisciplinary HSS department has been introduced. IDC, the research collaborations and industrial training Industrial Design Centre, celebrated 50 years opportunities help the students of IIT Bombay to of its golden existence earlier this year.
    [Show full text]
  • 1 Practical Training Cell, IIT Bombay
    Practical Training Cell, IIT Bombay 1 Director’s Message Prof. Devang Khakkar Director IIT Bombay IIT Bombay has proved to be the most pre- ferred destination for aspiring technologists from across the country. The institute con- sistently attracts the finest faculty and the best of students for its Bachelors, Masters and Doctoral Programs. IIT Bombay has a rich tradition of pursuing excellence and has con- tinually re-invented itself in terms of academic programs and research infrastructure. The presence of world class research facili- ties, institute industry collaborations, inter- national exchange programs, interdisciplinary research collaborations and industrial train- ing opportunities help the students of IIT Bombay to excel and be ahead in the com- petitive professional environment. We highly value our partnership with recruiters, alumni and friends of IIT Bombay and remain com- mitted to making your recruiting experience productive and positive. I invite recruiting or- ganizations and graduation students to find best match between their need and capabili- ties. Wish you all the best. 2 Practical Training Cell, IIT Bombay Dean’s Message Prof. Shiva Prasad Dean Academic Program IIT Bombay IIT Bombay offers perhaps the most flexible engineering curricula in the country to its students. The diverse skill of the student pop- ulation and the extensive theoretical knowl- edge they have will definately make them as assest to any organisation. At IIT Bombay, we believe that Practical Training or Internship is an integral portion of a student’s overall de- velopment and we encourage students to take up projects in industry and academia so that they can be better prepared for their future careers.
    [Show full text]
  • Visit to Stanford: a Report Dr
    Visit to Stanford: a Report Dr. Pushpak Bhattacharyya, Professor, Department of Computer Science and Engineering, IIT Bombay. [email protected] This summer (May-July, 2004), I visited Stanford University and other research organizations. Given below is a report of this visit. The description is divided into 7 parts: A. Visit to Stanford University. B. One day visit to Microsoft Research, Seattle. C. One day Visit to the Development Gateway Foundation, World Bank, Washington DC. D. One day visit to Honda Research Institute, Sunnyvale, California. E. Titles and Abstracts of talks delivered. F. Conclusions. G. The home page of the Stanford Natural Language Processing Group A. Stanford University The host institution for the visit was Stanford University. The objective of the visit was to set up collaborative research in Natural Language Processing and related areas between the research groups of IIT Bombay and Stanford University. The facts pertaining to this visit are as under: 1. Visitor: Dr. Pushpak Bhattacharyya, Professor, Computer Science and Engineering Department, Indian Institute of Technology, Bombay, India. 2. Place of Visit: Computer Science Department, Stanford University. 3. Duration: May-July, 2004. 4. Host and Principal Collaborator: Prof. Christopher Manning, Computer Science Department, Stanford University. 5. Research Group: Natural Language Processing Group of Prof. Manning. 6. Research group members interacted with (topics in parentheses): a) Roger Levy ("Probabilistic Parsing", "Psycholinguistic Experiments on Hindi Clauses"). b) Dan Klein ("Lexicalized and Unlexicalized Parsing"). c) Kristina Toutanova ("Word Dependencies", "PP Attachement"). d) Galen Andrews ("Implementation and Installantion of Stanford Parsers"). e) Teg Grenagar ("Probabilistic Context Free Grammar"). f) Iddo Lev ("Universal Networking Language", "Logic Puzzles").
    [Show full text]
  • 09.14-Bhattacharyya.Pdf
    Department of Computer Science University of Houston DISTINGUISHED LECTURER SEMINAR 2012 WHEN: FRIDAY, SEPTEMBER 14, 2012 WHERE: PGH 232 TIME: 3:00 PM SPEAKER: Dr. Pushpak Bhattacharyya, IIT Bombay Host: Dr. Ioannis Kakadiaris TITLE: Word Sense Disambiguation- a Multilingual Resource Conscious Perspective Abstract: Word Sense Disambiguation (WSD) is a fundamental problem in Natural Language Processing (NLP). Amongst various approaches to WSD, it is the supervised machine learning (ML) based approach that is the dominant paradigm today. However, ML based techniques need significant amount of resource in terms of sense annotated corpora which takes time, energy and manpower to create. Not all languages have this resource, and many of the languages cannot afford it. We discuss ways of doing WSD under resource constraint. We describe a novel scoring function and an iterative algorithm based on this function to do WSD. This function separates the influence of the annotated corpus (corpus parameters) from the influence of wordnet (wordnet parameters), in deciding the sense. Next we describe how the corpus of one language can help WSD of another language, i.e., LANGUAGE ADAPTATION. This is presented in three setting of "complete", "some" and "no" annotation. The talk is presented in a multilingual setting of Indian languages. The presentation is based on work done with PhD and Masters students and researchers: Rajat, Mitesh, Salil, Saurabh, Anup, Sapan and Piyush, and published in fora like ACL, COLING, EMNLP, GWC, ICON and so on. BIO: Dr. Pushpak Bhattacharyya is a Professor of Computer Science and Engineering at IIT Bombay. He received his B.Tech from IIT Kharagpur, M.Tech from IIT Kanpur and PhD from IIT Bombay.
    [Show full text]
  • Leveraging Cognitive Features for Sentiment Analysis
    Leveraging Cognitive Features for Sentiment Analysis Abhijit Mishray, Diptesh Kanojiay,|, Seema Nagar ?, Kuntal Dey?, Pushpak Bhattacharyyay yIndian Institute of Technology Bombay, India |IITB-Monash Research Academy, India ?IBM Research, India yfabhijitmishra, diptesh, [email protected] ?fsenagar3, [email protected] Abstract e.g., finding appropriate senses of a word given the context (e.g., His face fell when he was dropped Sentiments expressed in user-generated from the team vs The boy fell from the bicycle, short text and sentences are nuanced by where the verb “fell” has to be disambiguated) (3) subtleties at lexical, syntactic, semantic Domain Dependency, tackling words that change and pragmatic levels. To address this, polarity across domains. (e.g., the word unpre- we propose to augment traditional features dictable being positive in case of unpredictable used for sentiment analysis and sarcasm movie in movie domain and negative in case of un- detection, with cognitive features derived predictable steering in car domain). Several meth- from the eye-movement patterns of read- ods have been proposed to address the different ers. Statistical classification using our en- lexical level difficulties by - (a) using WordNet hanced feature set improves the perfor- synsets and word cluster information to tackle lex- mance (F-score) of polarity detection by ical ambiguity and data sparsity (Akkaya et al., a maximum of 3:7% and 9:3% on two 2009; Balamurali et al., 2011; Go et al., 2009; datasets, over the systems that use only Maas et al., 2011; Popat et al., 2013; Saif et al., traditional features. We perform feature 2012) and (b) mining domain dependent words significance analysis, and experiment on (Sharma and Bhattacharyya, 2013; Wiebe and Mi- a held-out dataset, showing that cognitive halcea, 2006).
    [Show full text]
  • Word Sense Disambiguation Algorithms in Hindi
    Word Sense Disambiguation Algorithms in Hindi Drishti Wali (13266) Nirbhay Modhe (13444) Department of Computer Science and Engineering, IIT Kanpur April 18, 2015 Abstract Word sense disambiguation (WSD) is the task of automatic identification of the sense of a polysemous word in a given context. Quite a lot of work has been done for WSD in English, inspired mainly by the Senseval and SemEval tasks. In this project we use the model of word representation stated in the paper by Chen, Liu and Sun3 for a Hindi corpus. This model represents the context and each sense of the target word as a vector in a high dimensional space and measures their similarity. The most similar sense is the chosen as the correct sense of the target word in the given context. 1 Contents 1 Motivation3 2 Resources and Corpus3 2.1 Hindi Wordnet..........................3 2.2 Hindi Corpus...........................3 3 Methodology3 3.1 Word2vec : Skip-Gram Model..................4 3.2 Sense and Context Representation...............4 3.3 Similarity using Cosine Distance................4 4 Results5 4.1 Threshold Analysis........................5 4.2 Example Cases..........................5 4.3 Insights..............................6 5 Future Work6 6 Conclusion6 2 1 Motivation Word sense disambiguation has lot of applications in improving search en- gines, machine translation, resolution of Anaphora etc. The first work in Hindi word sense disambiguation was done by Pushpak Bhattacharyya in 2004.8 Following this paper other works using bilingual methods and graph based approaches have been researched. The dearth of resources in Hindi has prevented the successful application of supervised algorithms in Hindi.
    [Show full text]
  • Machine Learning for Machine Translation
    Machine Learning for Machine Translation Pushpak Bhattacharyya CSE Dept., IIT Bombay ISI Kolkata, Jan 6, 2014 Jan 6, 2014 ISI Kolkata, ML for MT 1 NLP Trinity Problem Semantics NLP Trinity Parsing Part of Speech Tagging Morph Analysis Marathi French HMM Hindi English Language CRF MEMM Algorithm Jan 6, 2014 ISI Kolkata, ML for MT 2 NLP Layer Discourse and Corefernce Increased Semantics Extraction Complexity Of Processing Parsing Chunking POS tagging Morphology Jan 6, 2014 ISI Kolkata, ML for MT 3 Motivation for MT MT: NLP Complete NLP: AI complete AI: CS complete How will the world be different when the language barrier disappears? Volume of text required to be translated currently exceeds translators’ capacity (demand > supply). Solution: automation Jan 6, 2014 ISI Kolkata, ML for MT 4 Taxonomy of MT systems MT Approaches Data driven; Knowledge Machine Based; Learning Rule Based MT Based Example Based Statistical MT Interlingua Based Transfer Based MT (EBMT) Jan 6, 2014 ISI Kolkata, ML for MT 5 Why is MT difficult? Language divergence Jan 6, 2014 ISI Kolkata, ML for MT 6 Why is MT difficult: Language Divergence One of the main complexities of MT: Language Divergence Languages have different ways of expressing meaning Lexico-Semantic Divergence Structural Divergence Our work on English-IL Language Divergence with illustrations from Hindi (Dave, Parikh, Bhattacharyya, Journal of MT, 2002) Jan 6, 2014 ISI Kolkata, ML for MT 7 Languages differ in expressing thoughts: Agglutination Finnish: “istahtaisinkohan” English: "I wonder if
    [Show full text]
  • Webinar on 27 April by Prof Pushapak Bhattacharya: "Recent Trends in Chatbot : a Sentiment Perspective "
    Webinar on 27 April by Prof Pushapak Bhattacharya: "Recent trends in Chatbot : a sentiment perspective " Register for the session of Webinar on, " Recent trends in Chatbot : a sentiment perspective", to be presented on Monday 27 April 2020 at 11:00-12:30 IST on the behalf of Computer Science and Engineering Department, MNIT, Jaipur by Prof. Pushapak Bhattacharya, FNAE, Abdul Kalam Fellow, Director, IIT Patna. Note : You can also stream this webinar on your mobile device, including smartphone and tablet. The registration link is given below: Meeting link: Meeting link: https://mnitjaipur.webex.com/mnitjaipur/j.php?MTID=md357eba5d573704803a8791908555bef Meeting number: 580 162 228 Password: Chatbot Join by phone +91-11-6480-0114 India Toll (Delhi) +91-22-6480-0114 India Toll (Mumbai) Access code: 580 162 228 Abstract: The talk suggest how sentiment and emotion analysis is gaining increasing importance in dialogue systems whose practical applications are in the form of chatbots. The talk starts with a description involving the foundations of sentiment and emotion analysis and will proceed by explaining how to incorporate sentiments and emotions in these chatbots. Duration : 1.5 hours (including audience Q&A) Presenter Prof. Pushpak Bhattacharyya is a computer scientist and a Professor at Computer Science and Engineering Department, IIT Bombay. Currently, he is also the Director of Indian Institute of Technology Patna. Prof. Pushpak Bhattacharyya has completed his undergraduate studies from IIT Kharagpur (B. Tech.) and Masters from IIT Kanpur (M.Tech). He finished his Ph.D. from IIT Bombay in 1994. He is a past President of Association for Computational Linguistics (2016- 17), and Ex-Vijay and Sita Vashee Chair Professor.
    [Show full text]
  • Summer Workshop on Ontology, NLP, Personalization and IE/IR Inaugural
    Summer Workshop on Ontology, NLP, Personalization and IE/IR IIT Bombay conducted a Summer Workshop on Ontology, NLP, Personalization and IE/IR sponsored by HP Labs, Bangalore on 16-18th July, 2008. The workshop faculty consisted of renowned experts in their respective fields. The participants were students and faculty of academic institutions from all corners of the country and also researchers from top search engine and NLP industries like Yahoo, Rediff, IBM and MSR. 130 participants were selected from approximately 400 applicants from all over India with the following distribution: Students: 70 Faculty from academic institutes: 23 Professionals: 37 The Workshop addressed the cutting edge technology areas of: 1. Ontology (hierarchical knowledge organization) 2. Bridge between text and Ontology 3. Semantic Relation Extraction. 4. Mathematical aspects of Information Retrieval. 5. Graphical Models. 6. Web Corpora Processing. 7. Sentiment Analysis. 8. Cross-Lingual Search. 9. Summarization Inaugural Session The workshop was inaugurated with an invocation to Goddess Saraswati sung by Mr. Abhisekh Sankaran, a PhD student of the CSE, Depatrent, IIT Bombay. During the inaugural, Prof. Pushpak Bhattacharyya- workshop chair- introduced the workshop by underlining the importance and the relevance of the event in the context of the multilingual web. Prof. Vasi- Deputy Director IIT Bombay- mentioned that IIT Bombay was celebrating its golden jubilee this year, and a number of important events including lectures by Nobel Laureates were taking place in the institute. Prof. Ramamritham- Dean R & D, IIT Bombay- underlined the importance of doing interdisciplinary research in today’s world which sees immense growth and complexity of knowledge. Dr.
    [Show full text]