Information Retrieval System for IT Service Desk for Production Line Workers

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24310%2F21%3A00008910" target="_blank" >RIV/46747885:24310/21:00008910 - isvavai.cz</a>

Výsledek na webu

<a href="https://mme2021.v2.czu.cz/en/r-16791-news-mme-2021/proceedings-of-the-39-th-international-conference-on-mme-202.html" target="_blank" >https://mme2021.v2.czu.cz/en/r-16791-news-mme-2021/proceedings-of-the-39-th-international-conference-on-mme-202.html</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Information Retrieval System for IT Service Desk for Production Line Workers

Popis výsledku v původním jazyce

The information retrieval system gets answers to its users’ questions. As a part of an IT service desk available for the workers at the production line it recognizes a spoken request and selects a proper answer or action for this request. Speech recognition done at the beginning of the process and presentation of the selected answer done in the end of the process is solved using freely available modules for speech recognition and text to speech. The subject of the presented research is the process of selecting answers to recognized questions. The software solving this process is done by the authors of this contribution. The main result of the research is the comparison of two approaches of selecting answers to questions, namely decision tree and neural network. The input of both approaches is the set of keywords recognized in the user’s speech. Due to the lack of real data from a real production line, the training set of keywords and correct answers is generated artificially. The result of the research shows that the neural solution is advantageous when the number of possible answers and the number of keywords in questions is high.

Název v anglickém jazyce

Information Retrieval System for IT Service Desk for Production Line Workers

Popis výsledku anglicky

The information retrieval system gets answers to its users’ questions. As a part of an IT service desk available for the workers at the production line it recognizes a spoken request and selects a proper answer or action for this request. Speech recognition done at the beginning of the process and presentation of the selected answer done in the end of the process is solved using freely available modules for speech recognition and text to speech. The subject of the presented research is the process of selecting answers to recognized questions. The software solving this process is done by the authors of this contribution. The main result of the research is the comparison of two approaches of selecting answers to questions, namely decision tree and neural network. The input of both approaches is the set of keywords recognized in the user’s speech. Due to the lack of real data from a real production line, the training set of keywords and correct answers is generated artificially. The result of the research shows that the neural solution is advantageous when the number of possible answers and the number of keywords in questions is high.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

39th International Conference on Mathematical Methods in Economics (MME 2021) Conference Proceedings

ISBN

978-80-213-3126-6

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

349-354

Název nakladatele

Czech University of Life Sciences Prague, Kamýcká 129, Prague 6, Czech Republic

Místo vydání

Prague, Czech Republic

Místo konání akce

Czech University of Life Sciences Prague, Kamýck

Datum konání akce

1. 1. 2021

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

000936369700057