Faculty of Engineering, LTH

Denna sida på svenska This page in English

Software@LTH events

CS MSc Thesis Zoom Presentations 7 February 2022


From: 2022-02-07 08:15 to 15:00
Place: In Zoom (see separate link for each presentation
Contact: birger [dot] swahn [at] cs [dot] lth [dot] se
Save event to your calendar

Two Computer Science MSc thesis to be presented on 7 February

Monday, 7 February there will be two master thesis presentations in Computer Science at Lund University, Faculty of Engineering.

The presentations will take place in Zoom. A separate link is provided for each presentation.

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation ( Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions are found on this page.


Presenters: Simon Grimheden, Joel Järlesäter
Title: Concretizing CRISP-DM for Data-Driven Financial Decision Support Tools
Examiner: Elizabeth Bjarnason
Supervisors: Markus Borg (LTH), Johan Crafoord-Larsen (Hetch AB)

To support development of applications utilizing Artificial Intelligence (AI) and/or Machine Learning (ML), so called data-driven applications, development process models such as CRISP-DM have been created. However, previous papers on the topic of CRISP-DM have concluded that the model lacks detailed method recommendations, hindering its use for developers without previous knowledge in the field. In this paper, we contribute to this research by creating a detailed CRISP-DM model, tailored for the domain of data-driven financial decision support tools. To achieve this, we interviewed companies that would be potential stakeholders in the domain, as well as potential developers of an application within the domain. Our research found three main challenges when developing data-driven financial decision support tools and resulted in a detailed version of the CRISP-DM model. Our suggested concretization of CRISP-DM features a more holistic approach to evaluation, as well as concrete recommended activities for each phase of the original CRISP-DM model.

Link to presentation:

Link to popular science summary: TBU


Presenters: Oskar Wändesjö, Adrian Göransson
Title: Evaluating ClickHouse as a Big Data Processing Solution for IoT-Telemetry​​​​​​​
Examiner: Alma Orucevic-Alagic​​​​​​​
Supervisors: Markus Borg (LTH), Anton Friberg (Axis Communications)

With data analysis migrating into the realm of big data, conventional storage and analytics tools have begun to show their limits. To address the problems with greater volume, veracity, and variety of data, ClickHouse, among other new technologies, has emerged. ClickHouse is a column-oriented OLAP DBMS developed at Yandex, open-sourced in 2016, and in September 2021 was spun out, creating ClickHouse, Inc. reaching a valuation of two billion US dollars the following month. This thesis evaluated ClickHouse on telemetry data analysis through experimental benchmarks based on real use-cases at Axis Communications with real-life metrics from IoT devices. ClickHouse was compared to the current implementation comprising Elasticsearch and MinIO as an on-premise solution. The results established ClickHouse as a suitable candidate to handle the challenges of big data processing while still being cost-effective and highly approachable for new adopters.

Link to presentation:

Link to popular science summary:ändesjöGöransson.pdf