Organizing information is a key challenge in companies. Data is often available in unstructured formats, and searching through manuals, product catalogs, documentation or internal guidelines is tedious and time-consuming.

MicroNova relies on an AI-based solution to improve the company's internal knowledge management: this enables structured questions and answers to be created automatically based on defined sources and transferred to a knowledge database after a quality check.

1. Automated generation of questions and answers

Suitable knowledge sources are first defined as the context for the questions and answers and transmitted to the AI, for example the content of product data sheets or websites. Other sources, such as customer tickets or email histories, can also be consulted. This creates a comprehensive knowledge base that can be continuously expanded.

2. Quality assurance by internal product experts

To ensure the quality of the content generated, all questions and answers are checked and approved by internal expertsThis step guarantees that the questions are relevant and the content is correct.

3. Integration into a knowledge database

The validated data is then integrated into a knowledge database - a vector database, for example. This forms the basis for applications such as internal reference works, self-service portals or chatbots.

ChatGPT from Open AI serves as the technical basis. ChatGPT is integrated via Microsoft Azure. In order to meet German data protection requirements, the data is processed exclusively in the Swedish region. This ensures that the solution is GDPR-compliant and adheres to the highest security standards.

A practical use case in our company shows how the AI knowledge database can be used and what advantages it offers:

MicroNova provides information on around 80 ManageEngine products relating to IT management on the website manageengine.de. The website comprises around 1,000 subpages. We use AI to create ten questions and answers and associated metadata (product, category, etc.) for each of the relevant web pages.

One Example:

The request to the AI is: "Create a list of 10 questions and 10 answers that a user could ask about the following website. Output the answers in JSON array format."

For context, we provide the content of the subpage for the product " ADAudit Plus ", an IT security and compliance solution for Windows-based environments

The AI lists questions that can be answered with the information on the website, for example in extracts:

1. Question: "What is ADAudit Plus?"
   Antwort: Answer: "ADAudit Plus is a web-based IT security and compliance solution that IT administrators to monitor and audit changes in Active Directory in real time."
    
2. Question: "What functions does ADAudit Plus offer?"
   Answer: "ADAudit Plus offers features such as user login monitoring, audit trails for privileged users, anomaly detection, contextual event analysis and over 200 preconfigured reports."
    
3. Question: "How many editions of ADAudit Plus are there?"
   Answer: "ADAudit Plus is available in three editions: Free Edition, Standard Edition and Professional Edition."
    
4. Question: "How can I download ADAudit Plus?"
   Answer: "You can download ADAudit Plus from the official website by clicking on the download section."

The content created helps customer service to answer queries more efficiently and can also serve as the basis for self-service portals or FAQs. This solution is made possible by a cloud architecture and the use of open API interfaces.

The automated creation and validation of content offers many advantages:

• Greater efficiency: Automated processes save time and costs compared to manual creation.
• Increased precision: Validated content guarantees precise answers.
• Flexibility: Different data sources can be easily integrated.
• Cost efficiency: Generating 10,000 questions and answers costs approx. 0.70 euros

An AI-supported knowledge database creates a flexible, scalable basis for further digitalization projects and a wide range of possible applications:

• Internal reference works: Quick access to guidelines or technical documentation.
• External self-service portals: customers can find answers to frequently asked questions without having to wait.
• Advanced AI applications: Chatbots can access the database directly and provide more precise answers.

The automation of knowledge management processes through AI optimizes efficiency and quality. Companies benefit from reduced costs, better customer communication and a basis for future-proof technologies.