AI-based categorization and contract fidelity

Purpose and goal: ** Denna text är maskinöversatt ** The goal was to develop an existing AI-based system to expand its scope of use to be able to categorize

purchase data and measure contract fidelity. The core of the system is that we use an LLM that can interpret obscure invoice line descriptions. We can now connect different databases even if the unique purchase line is not exactly the same in its original format. We compile purchases and agreements that create the basis for various reports and analyzes in our Spend tool. The spending tool is the hub and interface for our customers.

Expected results and effects: ** Denna text är maskinöversatt **

The result has become a powerful tool for analyzing purchases for both the public sector and private companies. We can now offer a more complete solution to customers in aim to analyze their purchases and find discrepancies between actual purchases and agreed term. With this information, you can then streamline your purchases and in that way save both time and money.

Approach and implementation: ** Denna text är maskinöversatt **

We have started from an existing system and arrangement that we have worked with for several years. Thanks to the development of AI technology, we have been able to create an even more powerful tool that "cleans dirty data". With this original knowledge, we have been able to create a specification for how the system would be structured and together with system developers developed the new system.

The work has continued throughout the autumn with frequent contacts to ensure functionality and user-friendliness.