Intelligence and the Web

The web is growing very fast and in an unorganized manner. For users of the web it is difficult to find and explore provided information. Therefore, many of the intelligent web applications nowadays contain technologies which allow for intelligent information retrieval to locate and access relevant information for a user.

These information retrieval methods are not only of interest for web applications, but also for a wide range of other scenarios where relevant information has to be found in large collections of electronic documents. Intelligent filtering and recommendation techniques can be used to customize the provided information further towards an individual user’s interests.

Specific topics covered by the module are:

• Information retrieval, including document representation, retrieval methods, and evaluation of retrieval performance.
• Recommender systems, including content-based and collaborative filtering techniques.
• Knowledge representation and semantic web, including metadata concepts such as RDF Schema and XML topic maps.

After having followed this module, the participants should

• have knowledge about key concepts and techniques for knowledge representation, and
• be able to apply techniques for information retrieval and for constructing recommender systems.

Three full-day lectures combined with project work and distance-based studies.

Individual, internal, oral, graded examination based on a report documenting a miniproject carried out in connection with the module. The grade is based on the 7 grade scale.

This module introduces fundamental data-mining and machine learning techniques, and shows their applications in discovering knowledge from web-data such as usage logs, web content, and web structure. For example, by mining usage-logs we may adapt and customize web-pages to individual users or make targeted advertisement campaigns.

Specific topics covered by the module are:

• Fundamental data-mining techniques for data clustering and association rule mining.
• Machine Learning techniques for constructing models for prediction and decision support.
• Web data mining, including web usage, content, and structure mining.

During the module, software tools for data mining and machine learning will be
demonstrated.

After having followed this module, the participants should be able to

• analyze a web mining problem with the aim of selecting a suitable data mining or machine intelligence technique for solving the problem, and
• apply key data mining and machine intelligence techniques for analyzing and mining an existing collection of web data.

Three full-day lectures combined with project work and distance-based studies.

Individual, internal, oral, graded examination based on a report documenting a miniproject carried out in connection with the module. The grade is based on the 7 grade scale.

The module is conducted as project work, and constitutes the practical component of the course. The participants will be divided into smaller groups (possibly one person groups), and should focus on solving a specific problem in an industrial context using the techniques presented in one of the first two modules.

The project work should be documented in a report that forms the basis for the final evaluation.

This module will focus on the application of intelligent web techniques. The aim is to analyze, apply, and evaluate selected core techniques from one (or both) of the first two modules in an industrial context of the participant’s own choice.

After the project has been completed, the participants should be able to

• describe the application of core information retrieval or knowledge discovery techniques in relation to the chosen industrial context. The focus will depend on the type of the project.
• account for the theory and principles underlying relevant information retrieval or knowledge discovery techniques. The focus will depend on the type of the project.
• make a prototype implementation of the chosen techniques, possibly using existing software systems.

This module is conducted as project work.

Individual, external, oral, graded examination based on a project report documenting the project that has been conducted in connection with the module. The grade is based on the 7 grade scale.