TSI logo 25 years
LV EN
Home > About Us > Structure and Government > Abstracts

DEGREE

PROGRAMME

FACULTY

YEAR

LANGUAGE

KEYWORDS

COMPARATIVE ANALYSIS OF LLM-BASED APPROACHES FOR SQL GENERATION

The rapid development of Large Language Models has unlocked opportunities for restructuring software development processes in general as well as in such cases as converting natural language into SQL queries. This study seeks to experimentally evaluate the effects of four LLM-based methods on the efficiency and quality of SQL generation. Evaluation is being held based on following metrics: Correctness, Completeness and Consistency. Studied LLM-based SQL generation methods include Specific LLMs tailored for SQL code generation like SQL Coder frameworks for generating SQL code (Vanna.ai, 2023; Llamaindex, 2023) and Multi agent collaborative networks for transforming language into SQL.The research utilizes a mix of literature review case studies and simulations. It offers a comprehensive review of the advancements in LLM-driven SQL generation encompassing concepts, technologies, methodologies, strengths, limitations, and ethical considerations.This research successfully bridges the gap between theoretical foundations and practical application of AI-augmented approaches while promoting the integration of LLM-based SQL generation, into automated software development processes.

Author: Maksim Ilin

Supervisor: Dmitry Pavlyuk

Degree: Master

Year: 2024

Work Language: English

Study programme: Computer Sciences

More...

Web service development for the enterprise

The aim of the work is to develop a website for a car service in accordance with the functional and non-functional requirements of a company that previously did not have a personal website. This website should help the company attract new customers, as well as retain existing ones, thanks to a more detailed description of all services provided and a pleasant interface. One of the main goals of this work is: to examine and compare an internet site that was developed using the classical method and a website whose code was written using artificial intelligence, and to provide an assessment of certain results.To find out which features are useful and what customers really want to see on the site, a survey of potential customers was conducted.Analyzing the subject areas, web pages with similar functionality were analyzed and identified, as well as the initial requirements for the website.At the design stage, development tools were analyzed and selected, a preliminary page design was developed, use case diagrams, page navigation tree were created.Also, when the code was written, manual testing was done, as well as code testing in individual applications.After the work was completed and the code was written, certain tests and code reviews were performed. The code was then fixed.

Author: Radions Kasmausks

Supervisor: Boriss Mišņevs

Degree: Bachelor

Year: 2024

Work Language: Latvian

Study programme: Computer Science

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...

Application of time series algorithms for container imbalance forecasting using event data.

The research aim is to evaluate a time series models application for container imbalance forecasting using container event data. Compare the model results and conclude the real-world application options and limitations. The research object is time series models of container imbalance forecasting and subject is performance of models for container imbalance forecasting based on event data.There are three chapters of the research. The first one is State of the Art on Empty Container Repositioning (ECR) forecasting methods and approaches. The second part is investigation of container imbalance forecasting opportunities using event data. The third part is an application of time-series methods for forecasting container imbalance, experiments with real data and attempts to develop a novel data-driven framework for event data trained time series model evaluation. The third part consists of training experiment results analysis and interpretation. The 8 different models of ARIMA, VAR, VECM algorithms were tested and evaluated by different container size and type combinations, as well of 6 different port locations. Finally, the research conclusions are followed by references and attachments.

Author: Vjačeslavs Matvejevs

Supervisor: Dmitry Pavlyuk

Degree: Master

Year: 2024

Work Language: English

Study programme: Computer Sciences

More...

Table View
Text View