Vous êtes sur la page 1sur 3
25 JUN 2018. CEN) va aT rego TT TTT TT ETT Question Paper Code: 70156 M.B/M.Tech, DEGREE EXAMINATION, APRIL/MAY 2018 Elective ; Computer Science and Engineering P5094 — INFORMATION RETRIEVAL TECHNIQUES (Common to M.Tech. Information Technology) (Regulations 2017) Time : Three Hours Maximum : 100 Marks eS se pw a © . What is automatic Local and Global analysis ? . Define Meta learning, Answer ALL questions PART-~A (10x2=20 Marks) Define information retrieval. . What are the evaluation metrics used for information retrieval ? List and define information retrieval models. What is a Structured Text Retrieval model ? Define Inverted indices. What is Flat Clustering ? Mustrate Dynamic Ranking, What is Link analysis ? PART-B (5*13=65 Marks) a) i) Explain the issues of Information Retrieval in retail market environment. (9) ii) Write short notes on Boolean Retrieval techniques. (a) (OR) ») i) Explain about the components of Web Search in detail. @ ii) Brief about Open Source Information Retrieval Systems. (6) 70156 ne & (A 12, a) Write short notes on : : "aay i) Boolean Model ii) Vector Model. (OR) b) i) Explain in detail about binary independence model for Probability Ranking Principle (PRP). ; @ ii) Describe about how to estimate the quety generation probability for query likelihood model. : @ 18. a) i) Explain distributed and dynamic indexing in detail. * : om if) Explain in deiail about dictionary compression. Oy (OR) 7 b) Explain in detail about the Rocchio algorithm for relevance feedback. 3) 14. a) Explain the following Extensions to the support vector machine model. (13) ’) Soft margin classification . ii) Multiclass support veetor machines ~ ii) Nonlinear support vector machines. (OR) b) i) Brief about Hierarchical agglomerative clustering in detail: ii) Explain in detail about Latent semantic indexing, © 15. a) Explain about the various XML Retrieval model for Multimedia Information Retrieval in detail: i (13) OR) : b). Explain about Indexing and Searching in Parallel and Distributed Information q Retriéval in detail: 7 H (a3) 16. a) Given a data tuple having the values “systems”, “26 the attributes department, age and salary, respectively, what would a naive 3 PART-C 70156 (1x15=15 Marks) © 30” and “46-50K” for Bayesian élassification of the status for the tuple be ? (16), DEPT. [status COUNT sales senior 30 sales junior 40 sales junior 40 , ‘systems junior 20 systems. senior 5 ystems. juni 3. systems 3 marketing enior 10 marketing: [junior 4 secretary senior 4 secretary junior 6 : (OR) ; ! b) i): How would you create the dictionary in blocked sort-based indexing on the fly to avoid anextra pass through the data? i), Why is it better to partition hosts (rather than individual URLs) between the nodes of a distributed crawl system ? @) | @

Vous aimerez peut-être aussi