Identifying Variables that Improve Communication with Bots

Abstract

Some people find communicating with bots so difficult and daunting that they avoid interaction, as they do with the obnoxious friend or colleague who has the answer to their question. Eventually, they communicate with the colleague to obtain the answer that will benefit them and the success of their endeavor. As technology advances, more businesses and households are computerized with the Internet of Things, robots, chatbots, virtual assistants and other forms of artificial intelligence (AI). Businesses are spending billions of dollars on AI, so it is imperative that today's workforce learns to communicate with bots despite the difficulties faced. Shankar states that by 2021 retailers will spend 6 billion dollars on AI [1]. Retail is one of many sectors investing in AI systems, so the 6 billion dollars is only a snippet of the total investment in AI. The purpose of this study is to assess human communication with bots and identify variables that improve communication between humans and bots by analyzing data from chatbots. Variables were identified through a review of the literature, data visualization and text mining. In reviewing the literature, there were obvious elements that engineers could implement to improve communication between humans and bots from enhance natural language processing to incorporating more data. Also, the analysis of the text data identified variables humans can implement to improve communication with bots. The significance of this study is to improve human and bot communication, especially as more and more careers will require human and machine collaboration. This study will benefit individuals who have a difficult time communicating with bots, so they can excel and be prepared for the future.