TSI logo
LV EN
Home > About Us > Structure and Government > Abstracts

DEGREE

PROGRAMME

FACULTY

YEAR

LANGUAGE

Mobile robotic welding cell

This paper examines the market for robotization in welding production. It proposes the development of a portable robotic welding cell based on these insights. The work is divided into four sections.The first section explores the current market for robotic welding systems and evaluates the necessity of integrating such systems into welding production. It discusses advantages like increased efficiency and precision, alongside disadvantages such as high initial costs and maintenance.The second section outlines essential criteria for designing a mobile robotic welding cell. This includes constructing a technical diagram detailing the location and set of required components, defining necessary dimensional criteria, and creating a 3D model and structural diagram to illustrate component interactions.The third section involves selecting components that meet the established criteria. A functional diagram is constructed to demonstrate how these components work together to achieve the desired functionality of the mobile robotic welding cell.The final section develops an electrical circuit diagram for the connection of components in the control device. It demonstrates the logical connection of components and their signal interactions using the Ladder programming language in PLC.

Author: Saveliy Martynov

Supervisor: Aleksandrs Kraiņukovs

Degree: Bachelor

Year: 2024

Work Language: English

Study programme: Robotics

More...

Web Service Development with Integrated AI Assistant for Advertisements of Automobile Vehicles

New advancements in AI can be used to improve the quality of certain services. Sub-section of AI – GenAI is one of the methods to use for simplifying mundane or repetitive tasks. Examples of such methods include the generation of recommended automobiles based on user’s needs and search of existing automobiles advertised within the scope of the system where such methods are applied. This project explores the integration of GenAI, specifically ChatGPT by OpenAI, into a web service as a tool to provide personalized recommendations to users. The goal is simplifying the search process by offering intelligent, AI-powered suggestions based on users’ needs. Despite similar projects, the potential benefit of this system lies in its ability to reduce the time and effort involved in automobile market research, thereby offering users valuable insights to aid in the selection of the most suitable automobile. The web service, developed using Flask, incorporates back-end technologies such as SQLAlchemy for database management. This project intends to look at past approaches in the related field of appliance to improve the overall experience of a user.

Author: Sergejs Burlaka

Supervisor: Karina Kostjkina

Degree: Bachelor

Year: 2025

Work Language: English

Study programme: Computer Science

More...

Personal Information Management using Adaptive Information Systems

Nowadays people struggle to handle their personal information in an organized and efficient manner. In average individual consumes around 74Gb per day (Heim, 2017), which contributes to to information overload and limits the amount of information that the human mind can handle.Due to human uniqueness, all people have different information preferences. Adaptive Information Systems (AIS) is a relatively new direction of research on the crossroads of Information Science, Human-Computer Interaction and Artificial Intelligence. It is an alternative to the traditional "one-size-fits-all" approach in the development of Information Systems. Considering that Information overload is a user-personal problem, AIS allow involve an adaption process inside the information system activate reinforcement learning for building knowledge about the user’s behavior in information processing. As a next step collected knowledge AIS is used for performing self-adaptive changes inside the system with the final goal of reducing information overload.This study proposed a new way of handling and control information overload by using AIS, which is supposed to consider and adapt to user's personal experience in personal information processing.

Author: Sergejs Paškovskis

Supervisor: Boriss Mišņevs

Degree: Master

Year: 2024

Work Language: English

Comparative analysis of online casinos risks in Latvia

The bachelor thesis consists of 3 chapters, the aim of which is to develop a proposal to improve the operation of online casinos, based on the comparative analysis, The first chapter will help to understand theoretical aspects, factors, concepts and operating principles of online casinos.Second chapter includes selecting and comparing 3 popular online casinos in Latvia.The third chapter analyses the risks of online casino operation.

Author: Sergejs Storožuks

Supervisor: Marina Koževņikova

Degree: Bachelor

Year: 2024

Work Language: Latvian

Study programme: Business and Management

More...

Enhancement Strategies for Retrieval-Augmented Generation Systems

This thesis systematically explores the enhancement of Retrieval-Augmented Generation (RAG) systems within Large Language Models, emphasizing optimization of retrieval parameters and generation accuracy. We investigate optimal configurations in RAG systems, including chunk size and overlap percentages, top-k selection, query transformations, different retrieval methods, different LLMs, namely GPT-3.5-Turbo and GPT-4, discovering that a chunk size of 500 tokens generally offers the best performance. Vector search using cosine similarity emerges as the most effective retrieval method, significantly enhancing both context precision and recall across various tasks and knowledge bases. Experimentation within the CRUD-RAG framework demonstrates its applicability in diverse tasks from content creation to knowledge refinement. Our findings indicate that enhancements in retrieval settings can markedly improve the performance of RAG systems, making them more efficient and adaptable for complex information synthesis and retrieval tasks. These results affirm the potential of systematic enhancements to improve AI-driven language models in practical applications, contributing significant insights and practical approaches to the evolving landscape of RAG system research.

Author: Sigita Lapiņa

Supervisor: Dmitry Pavlyuk

Degree: Master

Year: 2024

Work Language: English

Study programme: Computer Sciences

More...

Table View
Text View