We provides services for designing intelligent chatbot systems using Generative AI technology and Large Language Models (LLMs) in order to enable users to achieve intelligent linguistic interaction between humans and computers using Natural Language Processing and Generative AI, and by taking advantage of advanced information retrieval tools based on Machine Learning techniques, which contributes to business improvement and the speed and effectiveness of information retrieval. In this context, this chatbot provides the ability of retrieve textual answers to frequently asked questions (FAQs) regarding a specific topic or the content of specific books or documents. In this respect, the user enters questions into this system in natural language and the chat bot retrieves answers to them in natural language based on the knowledge it has been trained on which is related to frequently asked questions about a topic or books/documents in Arabic or Kuwaiti dialect. Based on the users’s question and the knowledge the system has been trained on, such as FAQs and book/document content, the intelligent assistant system retrieves information related to the user’s question using Generative AI and Natural Language Processing and displays it to the user.
In the context of unilizing OpenAI’s artificial intelligence (AI) technologies in developing the educational process, our consulting firm designed a smart chat bot system for Kuwait International Law School (https://kilaw.edu.kw) using generative AI technology and large language models (LLMs) and integrated it into the school’s website and the school’s learning management system to enable faculty members and students at the school to achieve intelligent linguistic interaction between humans and computers using Natural Language Processing (NLP) and Generative AI technologies in classical Arabic and the Kuwaiti dialect, and by taking advantage of advanced information retrieval tools based on machine learning technologies, which contributes to improving the educational process.
In this context, these technologies were used in the following:
1- The smart assistant system (Chat Bot Assistant) for admission was designed, which provides the ability to obtain text answers to frequently asked questions (FAQs) related to admission to the school by entering questions into this system in natural language to obtain answers to them in natural language by training the system on knowledge related to frequently asked questions about admission to the college. This system was used on the admission portal page on the college website to enable students applying to the college to obtain answers to their frequently asked questions about admission.
2- This technology was also used in information retrieval by designing a smart assistant system to answer students’ questions about the scientific content of law books to benefit from its content in the educational process, and this was applied to the scientific content of the book “Explanation of the Kuwaiti Penal Code, Special Section” by Dr. Faisal Al-Kandari, a professor at Kuwait International Law School.
Generative AI for Test Bank Generation
Generative AI, such as OpenAI’s ChatGPT, has emerged as a transformative tool in education, enabling the efficient creation of test banks and exam questions. By training ChatGPT on textbooks and teaching materials, educators can harness its capabilities to generate high-quality, customized assessment content. This process not only saves time but also enhances the teaching and learning experience. The project, Generative AI for Test Bank Generation, was implemented by Prof. Salah Alnajem, Professor of Computational Linguistics and Natural Language Processing at Kuwait University and Founder and CEO of Information Age Consulting (Kuwait). He used this project to create test banks and exams for all of the courses that he teaches at Kuwait University. The project utilized Generative AI through training OpenAI’s ChatGPT on teaching materials and textbooks which are OCR-converted or saved digitally as soft copies, and through Prompt Engineering to achieve academic benefits for the teacher, teaching process, and students. By automating the time-consuming process of preparing assessments, the curated ChatGPTs allowed Prof. Salah, as an instructor, to concentrate on curriculum development, coaching, and instruction. It also reduced the amount of work involved in creating several copies of the same question for various classes or exams. The trained ChatGPT has also been used to generate on-the-fly questions that are given orally to the students who have special visual or hearing needs in a one-to-one session (student-to-teacher session). This project enhanced the quality of assessment through achieving consistency in question quality, style, and formatting, which is guaranteed by the trained ChatGPT. Besides, when creating questions or answer keys, it lowers the possibility of human error. In addition, the curated ChatGPTs can be scaled up to enable the creation of assessments for large courses or many subjects throughout the college.
