System Approach to the Combined Use of Large Language Models and Classical Models in Foresight Tasks

Background. Large Language Models (LLMs) and their associated agents have spread wide technology and represent a significant advancement in recent times. These state-of-the-art models expose valuable potential, but they are not devoid of restrictions, inefficiencies, and limits. This article investi...

Full description

Saved in:
Bibliographic Details
Main Authors: Володимир Савастьянов, Михайло Столяр
Format: Article
Language:English
Published: Igor Sikorsky Kyiv Polytechnic Institute 2024-12-01
Series:KPI Science News
Subjects:
Online Access:https://scinews.kpi.ua/article/view/315079
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850196234779754496
author Володимир Савастьянов
Михайло Столяр
author_facet Володимир Савастьянов
Михайло Столяр
author_sort Володимир Савастьянов
collection DOAJ
description Background. Large Language Models (LLMs) and their associated agents have spread wide technology and represent a significant advancement in recent times. These state-of-the-art models expose valuable potential, but they are not devoid of restrictions, inefficiencies, and limits. This article investigates the exploration of these constraints within specific domain areas and prediction problems as examples. Objective. The article highlights features offered by GPT-based models and compares the conclusions with classical methods of textual data analysis in classification tasks using the prediction methodology as an example. The purpose of the study is to develop a system approach to the combined use of traditional machine learning approaches as a practical alternative to LLMs in foresight tasks using the example of STEEP analysis, which provides an opportunity to obtain valuable information from textual data. Methods. The study is structured into four segments, each addressing distinct parts: Data Mining, text pre-processing using LLMs, text pre-processing utilizing Natural Language Processing (NLP) methods, and comparative analysis of results. Data Mining includes data collection and data pre-processing stages for train and test observations. For the utilization of LLMs, chains of thought techniques and prompt engineering were used. Results. Throughout this study, it was acknowledged that the LLMs can be used in combination with classical machine learning methodologies for domain-specific areas in STEEP analysis under Foresight tasks. The outcome revealed a model that was developed significantly faster and with less complexity compared to LLMs such as GPT and Mistral. Increasing the number of models employed leads to more stable results. Conclusions. The main result of the proceeding is that the patterns that reveal LLMs under certain settings can also be identified by classical models. Moreover, augmenting the deployment of LLMs during the data preparation stages contributes to heightened stability in outcomes. Using classical models combined with LLMs speeds up response times during inference and reduces operating costs for running models.
format Article
id doaj-art-e7039e01d60b48f588c31eab8530bf25
institution OA Journals
issn 2617-5509
2663-7472
language English
publishDate 2024-12-01
publisher Igor Sikorsky Kyiv Polytechnic Institute
record_format Article
series KPI Science News
spelling doaj-art-e7039e01d60b48f588c31eab8530bf252025-08-20T02:13:31ZengIgor Sikorsky Kyiv Polytechnic InstituteKPI Science News2617-55092663-74722024-12-011371-410.20535/kpisn.2024.1-4.315079353639System Approach to the Combined Use of Large Language Models and Classical Models in Foresight TasksВолодимир Савастьянов0https://orcid.org/0000-0002-2052-0420Михайло Столяр1https://orcid.org/0009-0009-3624-3147National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, IASA, MMSANational Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, IASA, MMSABackground. Large Language Models (LLMs) and their associated agents have spread wide technology and represent a significant advancement in recent times. These state-of-the-art models expose valuable potential, but they are not devoid of restrictions, inefficiencies, and limits. This article investigates the exploration of these constraints within specific domain areas and prediction problems as examples. Objective. The article highlights features offered by GPT-based models and compares the conclusions with classical methods of textual data analysis in classification tasks using the prediction methodology as an example. The purpose of the study is to develop a system approach to the combined use of traditional machine learning approaches as a practical alternative to LLMs in foresight tasks using the example of STEEP analysis, which provides an opportunity to obtain valuable information from textual data. Methods. The study is structured into four segments, each addressing distinct parts: Data Mining, text pre-processing using LLMs, text pre-processing utilizing Natural Language Processing (NLP) methods, and comparative analysis of results. Data Mining includes data collection and data pre-processing stages for train and test observations. For the utilization of LLMs, chains of thought techniques and prompt engineering were used. Results. Throughout this study, it was acknowledged that the LLMs can be used in combination with classical machine learning methodologies for domain-specific areas in STEEP analysis under Foresight tasks. The outcome revealed a model that was developed significantly faster and with less complexity compared to LLMs such as GPT and Mistral. Increasing the number of models employed leads to more stable results. Conclusions. The main result of the proceeding is that the patterns that reveal LLMs under certain settings can also be identified by classical models. Moreover, augmenting the deployment of LLMs during the data preparation stages contributes to heightened stability in outcomes. Using classical models combined with LLMs speeds up response times during inference and reduces operating costs for running models.https://scinews.kpi.ua/article/view/315079system analysis; foresight; textual analytics; classification; llm; nlp
spellingShingle Володимир Савастьянов
Михайло Столяр
System Approach to the Combined Use of Large Language Models and Classical Models in Foresight Tasks
KPI Science News
system analysis; foresight; textual analytics; classification; llm; nlp
title System Approach to the Combined Use of Large Language Models and Classical Models in Foresight Tasks
title_full System Approach to the Combined Use of Large Language Models and Classical Models in Foresight Tasks
title_fullStr System Approach to the Combined Use of Large Language Models and Classical Models in Foresight Tasks
title_full_unstemmed System Approach to the Combined Use of Large Language Models and Classical Models in Foresight Tasks
title_short System Approach to the Combined Use of Large Language Models and Classical Models in Foresight Tasks
title_sort system approach to the combined use of large language models and classical models in foresight tasks
topic system analysis; foresight; textual analytics; classification; llm; nlp
url https://scinews.kpi.ua/article/view/315079
work_keys_str_mv AT volodimirsavastʹânov systemapproachtothecombineduseoflargelanguagemodelsandclassicalmodelsinforesighttasks
AT mihajlostolâr systemapproachtothecombineduseoflargelanguagemodelsandclassicalmodelsinforesighttasks