Mumbai University · Final Year Computer Engineering · Rev-2019 'C' Scheme
Big Data Analytics
CSC702 · Semester VII · 3 credits · Internal tests 20 (average of two) + end-semester 80
End-semester paper (3 hours): Six questions of 20 marks; Q1 is compulsory and you answer four.
Module 1: Big Data and Hadoop
2 hoursCharacteristics and types of big data; traditional versus big data approaches; case studies; Hadoop, its core components and ecosystem.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Characteristics and types of big data
- Hadoop and its ecosystem
Module 2: HDFS and MapReduce
8 hoursPhysical organisation of compute nodes; large-scale file systems; map tasks, grouping by key, reduce tasks and combiners; node failures; matrix–vector multiplication and relational algebra by MapReduce; Hadoop's limits.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Distributed file systems and HDFS
- MapReduce: map, shuffle and reduce
- Combiners and coping with node failures
- Matrix-vector multiplication by MapReduce
- Relational algebra operations by MapReduce
Module 3: NoSQL
10 hoursNoSQL and its business drivers; key–value, graph, column-family and document stores; shared-nothing analysis of big data; master–slave versus peer-to-peer distribution.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- NoSQL and its business drivers
- NoSQL data architecture patterns
- Key-value and document stores
- Column-family and graph stores
- Shared-nothing architecture and distribution models
Module 4: Mining Data Streams
11 hoursThe stream data model and DSMS; stream sources and queries; sampling streams; Bloom filters; counting distinct elements with Flajolet–Martin; counting ones in a window with DGIM; decaying windows.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- The stream data model
- Sampling data in a stream
- Bloom filters
- Flajolet-Martin algorithm
- DGIM algorithm
- Decaying windows
Module 5: Real-time Big Data Models
4 hoursRecommendation systems: content-based and collaborative filtering; product recommendation; social networks as graphs; clustering social graphs and finding communities.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Content-based recommendation
- Collaborative filtering
- Community detection in social graphs
Module 6: Data Analytics with R
4 hoursR, RGui and RStudio; expressions, variables and vectors; objects and scripts; reading, exporting and manipulating data; built-in functions; plots and visualisation.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Basics of R
- Data handling in R
- Data visualisation in R
Module contents are summarised from the official Mumbai University syllabus . AiCanCode.org is not affiliated with the University of Mumbai.