Analysing AI Chatbot API Functionality
Van Dijck, Jorn (2024)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2024052415652
https://urn.fi/URN:NBN:fi:amk-2024052415652
Tiivistelmä
The purpose of this thesis was to analyse and compare different chatbot platforms in terms of their response time, accuracy, multi-language support, ease of integration, and quality of
documentation. The research questions focused on understanding the strengths and limitations of chatbots and providing insights into their suitability for various applications. The goal was to provide a guide for system planning.
The thesis utilises a practical approach to conduct tests and evaluate the chatbot platforms. First, chatbots are introduced and explained. Next, the thesis discusses the workings of APIs and a short chapter about solutions without chatbots. Data analysis was carried out through quantitative and qualitative methods, examining the performance of each platform against predefined criteria.
The findings of the research demonstrate significant variations among the chatbot platforms in terms of their performance and usability. ChatBot showed a unique delay mechanism that
influenced its response time but provided a more humanlike conversation. Wit.ai demonstrated reliable accuracy and multi-language support, while GPT showcased advanced language capabilities and context understanding. The analysis indicates that each platform has distinct strengths and weaknesses, emphasizing the importance of selecting the correct platform based on specific project requirements.
While chatbots offer numerous benefits for automating repetitive tasks and enhancing customer interactions, they may not always be the best solution. It is essential to create a balance between the capabilities of chatbots and the human touch required for personalized and nuanced interactions.
documentation. The research questions focused on understanding the strengths and limitations of chatbots and providing insights into their suitability for various applications. The goal was to provide a guide for system planning.
The thesis utilises a practical approach to conduct tests and evaluate the chatbot platforms. First, chatbots are introduced and explained. Next, the thesis discusses the workings of APIs and a short chapter about solutions without chatbots. Data analysis was carried out through quantitative and qualitative methods, examining the performance of each platform against predefined criteria.
The findings of the research demonstrate significant variations among the chatbot platforms in terms of their performance and usability. ChatBot showed a unique delay mechanism that
influenced its response time but provided a more humanlike conversation. Wit.ai demonstrated reliable accuracy and multi-language support, while GPT showcased advanced language capabilities and context understanding. The analysis indicates that each platform has distinct strengths and weaknesses, emphasizing the importance of selecting the correct platform based on specific project requirements.
While chatbots offer numerous benefits for automating repetitive tasks and enhancing customer interactions, they may not always be the best solution. It is essential to create a balance between the capabilities of chatbots and the human touch required for personalized and nuanced interactions.