For many clients, we have seen a behavior where most users are only providing keywords when querying documents with our knowledge base MRC solutions. While this approach may sometimes lead to the correct answer, it is not the most effective way to use our tool, Alli.
Using full sentences when searching for information can provide more accurate and relevant results compared to using keywords. This is because full sentences provide a clearer context for what information is being sought, which helps our AI models to better understand the intent of the search. On the other hand, keywords can be ambiguous and may not provide enough context to identify the specific information being sought. By using full sentences, you can also filter out irrelevant results and obtain more targeted information.
When searching for information within Alli GPT, using full questions instead of keywords can improve the accuracy and relevance of your search results. By following these best practices, you can find the information you need quickly and efficiently.
This chart provides more information into the number of results i.e. question distribution between using open-ended questions versus a simple keyword search. The number of results and accuracy of the results both increase when using proper open-ended questions.
This can help you understand the difference in outcomes and further reinforce the importance of using full questions.
We hope that these tips will help you improve your search results and find the information you need quickly and efficiently. If you have any questions or would like to learn more about Alli, please don't hesitate to contact us. We are always happy to help.
Team of experts