SylabUZ
Course name | Web mining |
Course ID | 11.3-WE-BizElP-EkspZasInter-Er |
Faculty | Faculty of Computer Science, Electrical Engineering and Automatics |
Field of study | E-business |
Education profile | practical |
Level of studies | First-cycle Erasmus programme |
Beginning semester | winter term 2022/2023 |
Semester | 5 |
ECTS credits to win | 4 |
Course type | obligatory |
Teaching language | english |
Author of syllabus |
|
The class form | Hours per semester (full-time) | Hours per week (full-time) | Hours per semester (part-time) | Hours per week (part-time) | Form of assignment |
Lecture | 15 | 1 | - | - | Credit with grade |
Project | 30 | 2 | - | - | Credit with grade |
To familiarize students with basic models and techniques for discovering information found on the Internet
To familiarize students with text mining algorithms
Developing skills of exploring Internet resources based on statistical software.
Basics of statistics
Types of information on the internet. Introduction to Text Mining. Searching textual information. Preprocessing of text documents: removing unnecessary elements from text documents (stop list, punctuation, numbers, etc.), reducing words to the form of a semantic core using Porter's algorithm and selected IT libraries. Search by keywords. Organization of documents in the form of a term-document matrix (TDM) and various ways of calculating the weight of individual terms (TF - term frequency, IDF - inverse document frequency). Measures of similarity of vectors and using them to create a ranking of found documents. Comparing the quality of text document search engines using various measures, e.g. precision-recall, ROC curves. Selected elements of linear algebra and applying them to the task of TDM matrix approximation (Low-rank approximation), discussing the benefits of approximation. Various techniques for grouping and classifying documents. Document ranking based on connection structure: PageRank algorithm; authorities and hubs. Creating document summaries by automatically selecting the most important sentences and the most important words (terms). Creating wordclouds. Sentiment analysis as a technique to systematically identify, extract, quantify, and study affective states and subjective information (e.g. positive, negative, neutral, etc.). Word embeddings. Recommendation systems (user-based, item-based). Presentation of selected IT tools for carrying out tasks in the field of Text Mining.
Lecture, individual projects.
Outcome description | Outcome symbols | Methods of verification | The class form |
Lecture – the passing condition is to obtain a positive mark from the final test
Project– the passing condition is to obtain a positive mark from the project form
Calculation of the final grade: lecture 50% + project 50%
Modified by dr hab. inż. Artur Gramacki, prof. UZ (last modification: 21-04-2022 00:07)