SGBAU · Third Year B.E. Computer Science and Engineering (Data Science) · NEP 2020
Big Data Analytics
6DS234PE2 · Elective II · Semester VI · 3 credits · IE 40 + ESE 60
One of the Elective II options.
Unit 1: Big Data Analytics and Lifecycle
6 hoursBig data and the state of analytics practice, key roles in the big-data ecosystem, examples; the data-analytics lifecycle — discovery, data preparation, model planning and building, communicating results and operationalising — with a case study.
Examined in the 60-mark end-semester paper
- Characteristics of big data
- Classification of analytics
- Data analytics life cycle
- Big data applications
Unit 2: Analytics Methods, Clustering and Association Rules
6 hoursExploratory data analysis, hypothesis testing, difference of means, the Wilcoxon rank-sum test, type I and II errors and ANOVA; k-means clustering and its diagnostics; association rules, the Apriori algorithm, rule evaluation and frequent-itemset generation.
Examined in the 60-mark end-semester paper
- Exploratory data analysis
- Hypothesis testing basics
- Type I and type II errors
- Analysis of variance (ANOVA)
- K-means for big data
- Association rules and Apriori
Unit 3: Regression and Classification Techniques
6 hoursLinear and logistic regression models and their diagnostics, decision trees and their evaluation, the naive Bayes classifier, Bayes' theorem and smoothing, and diagnosing classifiers.
Examined in the 60-mark end-semester paper
- Linear and logistic regression
- Building a predictive model
- Decision tree learning and overfitting
- Bayesian learning and naive Bayes
- Cross-validation and model selection
Unit 4: Time Series and Text Analytics
6 hoursTime-series analysis and the Box–Jenkins method, AR, MA, ARMA and ARIMA models and the autocorrelation function; the text-analytics process, text representation, TF-IDF, topic categorisation and sentiment analysis.
Examined in the 60-mark end-semester paper
- Autocorrelation and partial autocorrelation
- ARIMA forecasting
- TF-IDF and vector space models
- Sentiment analysis
Unit 5: Big Data Tools: Hadoop and MapReduce
6 hoursHigh-performance architectures and storage, HDFS, the MapReduce model and YARN, ZooKeeper, HBase, Hive, Pig and Mahout, parallelism, developing MapReduce applications, and NoSQL in the Hadoop ecosystem.
Examined in the 60-mark end-semester paper
- Big data technologies
- Hadoop and MapReduce in the cloud
- Composing and incremental MapReduce
- HBase and CouchDB
- Apache Hive and HiveQL
- Apache Pig and Pig Latin
- Machine learning on big data: Mahout and Spark MLlib
Unit 6: Database Analytics, NoSQL and Graph Analytics
6 hoursAdvanced SQL and in-database text analytics; NoSQL and schemaless models — key-value, document, column and graph stores; graph models, triples, use cases and algorithms; the basics of visualisation and operationalising analytics projects.
Examined in the 60-mark end-semester paper
- NoSQL and its business drivers
- Schemaless databases and materialised views
- Key-value and document stores
- Column-family and graph stores
- Big data visualisation techniques
Unit contents are summarised from the official SGBAU syllabus . AiCanCode.org is not affiliated with Sant Gadge Baba Amravati University (SGBAU).