Test automation for named entity recognition system
Karsikas, Katja (2018)
2018
Tietotekniikka
Tieto- ja sähkötekniikan tiedekunta - Faculty of Computing and Electrical Engineering
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2018-11-07
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201810242443
https://urn.fi/URN:NBN:fi:tty-201810242443
Tiivistelmä
Software testing is an essential part of the software development process. It is needed to ensure the quality of software. As the software development process is changing towards continuous integration and deployment, an increase in the level of automation in testing is needed to ensure software quality all the time.
Using artificial intelligence (AI) creates additional challenges for testing. For instance, it can be challenging to determine whether the test output is unequivocally correct or not. In some cases, the tester can determine that AI software works correctly by using his own judgment or by getting the information from the product owner or the customer. However, in more challenging testing tasks with test automation, it needs to be very accurately specified what the correct output is and what is not.
This thesis is about implementing test automation for a named entity recognition system that recognizes person and organization names from text. Named entity recognition is a subcategory of AI. It means aiming to find and classify named entities in text into pre-defined categories like person and organization names. In cases of that kind, the correct outputs can be determined by people.
Using artificial intelligence (AI) creates additional challenges for testing. For instance, it can be challenging to determine whether the test output is unequivocally correct or not. In some cases, the tester can determine that AI software works correctly by using his own judgment or by getting the information from the product owner or the customer. However, in more challenging testing tasks with test automation, it needs to be very accurately specified what the correct output is and what is not.
This thesis is about implementing test automation for a named entity recognition system that recognizes person and organization names from text. Named entity recognition is a subcategory of AI. It means aiming to find and classify named entities in text into pre-defined categories like person and organization names. In cases of that kind, the correct outputs can be determined by people.