Page 1 of 1

Module Code - Title:

CS5062 - DATA ANALYTICS

Year Last Offered:

2021/2

Hours Per Week:

Lecture

2

Lab

0

Tutorial

1

Other

3

Private

4

Credits

6

Grading Type:

N

Prerequisite Modules:

Rationale and Purpose of the Module:

This module is a core module for the MSc in Artificial Intelligence. This module introduces the elements of the data analytics workflow, including data cleaning, feature extraction and feature selection, predictive and descriptive modelling, and deployment of models. The role of data visualisation in the data analytics process is discussed as well. The module involves practice with a state-of-the-art data analytics software platform.

Syllabus:

1. Introduction to data analytics; relation between data mining, data analytics, data science; motivation behind data analytics; cross-industry standard process (CRISP-DM) for data mining; data analytics workflows. 2. Data pre-processing: feature extraction, data cleaning, handling missing data, methods for identifying outliers, data transformation. 3. Methods for feature selection: filter, wrapper and embedded methods. 4. Styles of machine learning for data mining: supervised vs. unsupervised learning, classification, numeric prediction, clustering, association learning. 5. Algorithms for building predictive and descriptive analytics models: a. Predictive modelling algorithms for classification and numeric prediction, such as OneR, ID3, C4.5, Naïve Bayes, k-NN, Prism, SVM, linear regression, logistic regression, Perceptron, Winnow. b. Descriptive modelling algorithms for clustering and association learning, such as k-means, apriori, max-miner. 6. Evaluation of predictive and descriptive analytics models: Holdout and cross-validation, cost-benefit analysis, user feedback. 7. Visual analytics: methodology and workflow. 8. Case studies in subdomains, such as sentiment analysis, item/service ranking recommendation, image classification, etc. 9. Practical use of data mining platforms for building data mining workflows and training predictive and descriptive analytics models.

Learning Outcomes:

Cognitive (Knowledge, Understanding, Application, Analysis, Evaluation, Synthesis)

At the end of this module, students will be able to: 1. Summarise the main elements of the data mining workflow. 2. Differentiate predictive from descriptive analytics in terms of methods and output. 3. Recognise and describe at least one algorithm in each of the four categories: classification, numeric prediction, clustering, association learning. 4. Construct data mining workflows with the use of data mining software for training of predictive and descriptive analytics models. 5. Analyse the results of machine learning algorithms. 6. Recognise the role of data visualisation in the data mining process.

Affective (Attitudes and Values)

At the end of this module, students will be able to: 1. Discuss the benefits of data mining for industry and society.

Psychomotor (Physical Skills)

N/a

How the Module will be Taught and what will be the Learning Experiences of the Students:

Research Findings Incorporated in to the Syllabus (If Relevant):

Prime Texts:

Witten, Frank, Hall (2017) Data Mining : Practical Machine Learning Tools and Techniques , Elsevier Science & Technology
North (2016) Data Mining for the Masses , CreateSpace Independent Publishing Platform

Other Relevant Texts:

Han, Kamber, Pei (2017) Data Mining: Concepts and Techniques , Elsevier Science & Technology

Programme(s) in which this Module is Offered:

Semester - Year to be First Offered:

Module Leader:

Nikola.Nikolov@ul.ie