Data Mining II

Introduce the participant to additional Data Mining techniques. The different stages of the Knowledge Extraction process will be covered as a decision support tool.

Objectives:

 Objective

• Introduce in the additional techniques of Data Mining. The different stages of the Knowledge Extraction process will be covered as a decision support tool.

 Specific objectives

• Analyze models based on solving classification and prediction problems.
• Know the characteristics of knowledge and its relationship in research and innovation.
• Use data mining techniques for the relevance and significance of useful information for the organization.
• Identify behavior patterns based on a data set and the selection of attributes.
• Identify common characteristics in groups based on the recorded values.
• Apply discovery knowledge to problems underlying your profession.

 Course methodology

The course will be taught under the workshop modality through face-to-face classes in which, in addition to seeing the theoretical topics, simple and concrete examples will be solved and analyzed. Theoretical-practical learning where the contents of the course will be developed exemplifying specific situations. Specific problems will be solved collaboratively, promoting discussion. At the end of each topic, additional readings and practical activities will be suggested that students must solve outside of class hours.

At the end of the module, the student will be able to:

• Understands the fundamental concepts of data mining.
• Generates information processing models based on a methodology.
• Manipulates tools to build and generate knowledge.

Duration: 40 hours

Curricular content

• Introduction. Obtaining knowledge from data. The KDD process. Phases of the knowledge extraction process. Relationship with other disciplines. Decision trees. (Monday, March 06, 2023, 4 hours)
• Data Preparation. metadata. Input information analysis. Statistical measures. Construction and analysis of graphic representations. Cleaning and transformation. Transformation and creation of attributes. Discretization and Numerization, Range Normalization, scaling and centering. Exploration through visualization and selection of data. Decision trees. (Monday, March 06, 2023, 4 hours)
• Data Mining Techniques. Pattern Extraction. Introduction. Tasks and Methods. Predictive and descriptive tasks. Supervised learning and unsupervised learning. Decision trees and Clustering. (Tuesday, March 07, 2023, 4 hours)
• Grouping Techniques. Cluster quality metrics. Types of grouping: Hierarchical, partitive and probabilistic. Measures of distance and connectivity. Grouping process. Partitive clustering. k-means algorithm. (Wednesday, March 07, 2023, 4 hours)
• Classification rules. Partition vs coverage. ZeroR, OneR, PRISM, PART and CN2 methods. Metrics of a rule: support, coverage, trust, interest and conviction. (Thursday, March 07, 2023, 4 hours)
• Association rules. Quality of the rules. A priori algorithm. Frequent item concept. A priori algorithm improvements: FP-Growth (Friday, March 10, 2023, 4 hours)

Once the activities and requirements for the development of the course have been fulfilled, the participants will receive the certificate of Approval of the blended course: " Data Mining II”, based on the provisions of the Regulations of the Department of Continuing Education, that is, that the participant obtains a minimum of 70% in the academic evaluation and 80% attendance at face-to-face classes.

Teachers of the University of Azuay.

mgt. Marcos Orellana Cordero.

Course sponsored by the University of Azuay within the framework of the teacher training program.

UDA campus

formacion.continua@uazuay.edu.ec