Offers related to "Machine learning"

Applicability of Recommender Systems

Recommendation systems automatically match online users with products, documents or in general, items. They can, for instance, increase the click-through rate, purchases or average basket size in online shops. Based on a set of specifics and requirements of a domain and with access to data, offline experimentations will be performed.

System Design for Group Decision Support

Groups of various nature (e.g., in a family or in a work team) are continuously facing the need to make decisions about common issues. For instance, a software team must decide which requirements to implement in the next version of the system or a group of friends may want to decide which excursion to do. We can evaluate the applicability of user profiling and recommendation techniques for supp...

Chat Bot

Bots are becoming popular in many applications to simplify the information access processes and support users. We have experience in creating bots for helping students to select the right meal for them in the university canteen. This technology can be adapted in other application domains.

Applicability of Classification Algorithms

Classification is a machine learning approach to automatically assign a category to data records, where the categories can be binary values, such as good/bad, or can consist of multiple values. The basis for training a classifier is a large enough that constitutes the ground-truth, i.e., information about how the data should be labeled. Data can be, for instance, documents, sensor measurements...

Next User's Action Prediction Study

In many web scenarios users of online systems repeatedly access the same application. For instance, a user may visit the homepage of a newspaper or a streaming video application every day. In these cases, it is useful to be able to predict, the next time the user will visit the application, what will be his/her action, i.e., what document will be browsed or what video will be played. Data anal...

Applicability of Process Mining

This activity consists of a pilot consultancy on process mining. Process mining is a mining technique that allows to create a process model using the logs of the running process as input, to reason over it (e.g., identify bottlenecks) and to propose how to improve it. This activity will address data preparation, identification of most suitable techniques, and an initial analysis of the company...

Process mining

Teaching activity on process mining, covering the basics of process mining, then different process mining techniques, and the entire lifecycle of a process mining project.

Visualization of software bugs

Software bugs occur at different rate depending on the characteristic of the software systems. Automatically collecting bugs from online or offline source (e.g., issue repositories) and visualising them over time help to understand the reliability of the system and to predict its future behaviour. We offer to develop a simple tool that mines the repository and visualises optimal curves represen...

Feedback/Request info