DATA MINING

Single discipline educational activity

Course Sheet

Course code:

DAMI19

Academic Year of enrolment:

2020

Year of supply:

2020/2021

Type of Course:

Related/Supplementary

Degree:

Course of 2nd Cycle Degree in DIGITAL MARKETING

Disciplinary Sector:

Computer Science

Language:

Italian

Location:

PESCARA

Credits:

Programme Year:

Professor and Collaborators:

AMATO Gianluca

Cycle:

Second four months

Hours of classroom activity:

Individual study hours:

171

Prerequisites:

Although no prerequisites are formally required, basic knowledge of descriptive statistic and some computer literacy are strongly recommended.

Objectives

A digital marketing manager needs information to be able to design an effective campaign. This information, in most cases, does not come from traditional market surveys, but must be derived from the analysis of data collected for the most diverse purposes: website logs, sales data, digital advertising services such as Google Analytics, etc ... Sometimes the bulk of raw data is so vast that it is necessary to resort to distributed computational mechanisms: we enter into the realm of big data.

Obviously, the professional figure that is normally in charge of this type of analysis is not the digital marketing manager but the data scientist. However, it is important that also the managerial roles have a basic preparation in the field of data knowledge extraction, both in order to communicate more effectively with the roles in charge of this activity, and because, in medium-small contexts, a specific role of data scientist may not be available and the manager could find himself in the position of having to make up for this lack. In any case, having a basic preparation in the field of data mining is useful to learn about the potential and limits of the tools you have available. The aim of the course is therefore to provide this basic knowledge, as detailed in the section on course content.

More specifically, and referring to the Dublin descriptors, at the end of the course the student will be able to

KNOWLEDGE AND UNDERSTANDING:
• explain clearly and correctly the main concepts of data analysis and data mining;
• accurately illustrate the behaviour of the main analysis algorithms.

APPLYING KNOWLEDGE AND UNDERSTANDING:
• perform simple data analysis using the KNIME software;
• extract information to support marketing strategies through data mining methods.

MAKING JUDGEMENTS:
• choose the most appropriate type of analysis based on the objectives and the types of data available;
• interpret the results obtained from the data analysis, without having to resort to external experts.

COMMUNICATION SKILLS:
• communicate with any data science expert using a correct and appropriate language;
• ommunicate with other managerial roles the results of data analysis in a simple but correct language, making use of understandable and easily interpretable graphics.

LEARNING SKILLS:
• read and understand basic and advanced level (but not research level) papers on data analysis methodologies;
• read and understand papers on the applications of data analysis techniques to support marketing strategies.

Contents

In this course we will study the process of knowledge extraction from databases, in particular the phases of data cleaning and pre-processing, data analysis and validation.

In the data analysis part, we will mainly consider techniques based on artificial intelligence and machine-learning, since statistical techniques are the subject of other courses in the same program of study. Theoretical lessons will be supplemented by practical exercises using the KNIME Analytics Platform software (https://www.knime.com/). During the practical exercises, we will present some case studies that are of particular interest for the Digital Marketing Manager, such as market basket analysis.

Extended Syllabus

In this course we will study the process of knowledge extraction from databases, in particular the phases of data cleaning and pre-processing, data analysis and validation.

More in detail, the course content will be as follows:

• The process of extracting knowledge from data: data cleaning and integration, selection and transformation, data mining, validation, presentation.
• Data: data types, data sources and big data, data pre-processing, similarity measures and dissimilarity.
• Classification analysis: decision trees, artificial neural networks, overfitting, classifier performance evaluation.
• Association analysis: frequent items, association rules, interesting measures for patterns.
• Clustering analysis: partitioning methods, hierarchical methods, density-based methods, cluster validation.

Recommended Bibliography

• Pang-Ning Tan, Michael Steinbach, Vipin Kumar. Introduction to Data Mining (2nd edition). Prentice Hall.

• Rosaria Silipo. KNIME Beginners Luck.

Methods of Provision

Conventional

Teaching Methods

The course includes 54 hours of lessons, one third of which must be taught in the computer lab.

Evaluation methods

Verification of learning:

The assessment of the students' learning level will be carried out by combining a practcal and written test.

1) The practical test consists of a data analysis project of an intermediate level. Students will carry out the project by working, alone or in small groups, at home or in the computer lab,. All students will subsequently discuss the project individually with the teacher. At the end of the discussion, a mark will be assigned ranging from 0 (insufficient) to 4 (excellent). Students with an insufficient mark will have to repeat the practical test, possibly with a different project. The objective of the practical test is to evaluate the student's ability to complete a data analysis task according to the methodology studied in class. Therefore, in relation to the Dublin Descriptors, the practical test is mainly aimed at assessing the "applied knowledge" and the "making judgment" proficiencies

2) Students who have passed the project can take the written test, which consists of exercises and open-ended questions. The maximum mark of the written test is 30, while the minimum to pass the test is 18. The written test is aimed at assessing the student's acquired knowledge, mastery of the specific technical language of the subject treated and clarity of exposition.

The final mark will be the sum of the mark of the written test and the mark of the project. If the sum exceeds 32, the teacher will evaluate whether to award the honors.

The examination procedures are the same for attending and non-attending students.

Contacts/More Information

Further information is available on the course website in the university e-learning platform.

Università degli Studi
"G. d'Annunzio"

Chieti - Pescara

Aree

DATA MINING

Course year: 1

Course year: 2

Search form

Aree

DATA MINING