SGBAU · Third Year B.E. Computer Science and Engineering · NEP 2020
Big Data Analytics
6CS234PE2 · 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 and types of big data
- Big data stack and enabling technologies
- The data analytics lifecycle
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
- Data wrangling and exploratory data analysis
- Sampling, estimation and hypothesis testing
- ANOVA
- Clustering: K-means and hierarchical
- Association rules: Apriori and FP-growth
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
- Regression: linear and logistic
- Classification: naive Bayes and decision trees
- Model evaluation: confusion matrix, precision, recall and ROC-AUC
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
- Moving averages and ARIMA
- Time series decomposition
- Text analytics: bag of words and TF-IDF
- 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
- Hadoop and its ecosystem
- Distributed file systems and HDFS
- MapReduce: map, shuffle and reduce
- Apache Spark and YARN
- Apache Hive and HiveQL
- Apache Pig and Pig Latin
- ZooKeeper and Paxos
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
- Key-value and document stores
- Column-family and graph stores
- Graph processing: Pregel, Giraph and GraphX
Unit contents are summarised from the official SGBAU syllabus . AiCanCode.org is not affiliated with Sant Gadge Baba Amravati University (SGBAU).