Download PDF

　 (Head Office: Chiyoda-ku, Tokyo; CEO: Osamu Hata; hereinafter referred to as "WAP") will release a new feature of its AI-type chatbot "HUE Chatbot" that supports RAG (Retrieval-Augmented Generation)*1 in July 2024. This feature will be released in July 2024. This feature will allow AI to automatically generate highly accurate responses by referencing documents specific to each company as specified by the chatbot's administrator.

　This is expected to not only significantly reduce the man-hours required to respond to inquiries and provide information within the company, but also improve operational efficiency and productivity by enabling users to quickly obtain the information they need.

1. Generation AI automatically cites the necessary elements Answer More readable

　In order to automatically answer various questions using natural language processing (NLP), chatbot administrators need to set up and update FAQs. In order to reduce the burden on the administrator, "HUE Chatbot" has a "Document Search Option" that allows the administrator to search for uploaded document files such as company rules and manuals, or specified websites, and the chatbot will provide answers as they appear in the search results. The search results are then presented as answers to the questions.

　Now that this option is compatible with RAG, Microsoft's Azure OpenAI Service will refer to documents related to the question and use AI to generate the most appropriate answer, enabling more accurate answers in the form of text.

2. All initial development has been handled.

　The "Document Search Option" of HUE Chatbot supports RAG as a standard feature, so there is no need for development required for RAG implementation, such as search preprocessing or prompt construction.

No development man-hours required for normal RAG implementation

*1: "Retrieval Augmentation Generation (RAG) is an architecture that extends the functionality of a Large Language Model (LLM), such as ChatGPT, by adding an information retrieval system that provides grounding data. Adding an information acquisition system gives you control over the grounding data used by the LLM when creating a response."
Source: Microsoft, Microsoft Build,: https: //learn.microsoft.com/ja-jp/azure/search/retrieval-augmented-generation-overview

What is HUE Chatbot?

　An AI-type chatbot that responds to ""fluctuations"" in Japanese by applying NLP (natural language processing) technology. Equipped with "SudachiDict," Japan's largest natural language processing dictionary with a registered vocabulary of 2.9 million words (as of May 2024), it accurately judges the unique fluctuations of Japanese notation and responds to ambiguous questions, enabling highly accurate dialogue as if it were talking to a person. In addition, the usage status and points for improvement are visualized on a dashboard, allowing anyone to perform PDCA cycles with intuitive operation.

• HUE Chatbot 
• Works Tokushima Artificial Intelligence NLP Laboratory 

Demonstration of ERP x Generative AI on July 18 with Microsoft Japan Co.

　At "WorksWay 2024," an online business forum held by WAP, Microsoft Japan, Inc. and WAP's General Manager will talk about the innovations brought by the combination of ERP (Enterprise Resource Planning) and generative AI, and discuss Microsoft Copilot for Microsoft 365 and the latest technologies offered by HUE will be demonstrated.

Click here to register for "WorksWay 2024".

About Works Applications

Works Applications was founded in 1996 as an ERP package vendor in Japan. With innovative solutions such as no-customization and free version upgrades, we have supported the growth of our customers, mainly major Japanese companies. Believing in the potential of each individual, we aim to be a "growth engine" that maximizes the value of companies and individuals, and we will continue our pursuit of turning "work" into "creation" and making "work" fun.

*Microsoft, Microsoft Build, Azure, and Microsoft 365 are either registered trademarks or trademarks of Microsoft Corporation in the United States and/or other countries.
*Company names, product names and service names are trademarks or registered trademarks of their respective companies.
*The information in this release is current as of the date of publication, and is subject to change or withdrawal without notice. Please be aware that any forecasts or forward-looking statements in this release are based on uncertainties and may differ from actual results.