In this blog, I’ll share the high level differences between Microsoft’s Power BI Q&A and customized AI Chatbot solution that might help if you can’t decide which of the above is suitable for your particular use case.

To know about Power BI Q&A feature, follow the link below: https://docs.microsoft.com/en-us/power-bi/consumer/end-user-q-and-a

Both Power BI Q&A and chatbot supports answer to natural language queries and are customizable for any specific domain and user requirements but both have their individual strengths and capabilities. In the below paragraph, I’ll share the key features of both of them and compare accordingly.

When and what to choose ?

Consider the below aspects before deciding which one out of both Power BI Q&A and chatbot can fulfill your requirements:

The most important aspect to consider is the development time. To make a customized chatbot from scratch requires time and maintenance whereas Power BI Q&A feature is readily available on dashboards and report on Power BI service. Also, Power BI has added Q&A capability to Power BI mobile apps also so it is easily accessible

In Power BI Q&A you can import data in the form of text, csv, excel, database etc. directly into dashboard whereas you need to train your chatbot on various types of questions that can be possibly asked in different linguistic patterns. It takes time and effort to prepare training data accordingly.

Power BI Q&A supports natural language queries in English only whereas chatbot can be trained and developed on different languages

Q&A feature answers to queries by looking the keywords in the dataset or configuration associated with the dashboard and report. If any one of the keywords in query doesn’t match with the column or values in dataset,it will miss out that column from the response whereas in chatbot, it gives user flexibility to type query in any pattern or format and it will classify the intent and entities using ML models

In addition to the above point, to use Power BI Q&A feature, user should be well informed about the data (columns and values) in dashboard so that he/she can phrase query in a way that keywords could be picked up by Power BI

The output response in Power BI doesn’t provide any business justification that how response was derived from data or how calculations were done behind the scene whereas the justification part can be customized and developed to be answered after each query in chatbot

The functionalities in Power BI Q&A like Visualizations , keyword suggestions, adding synonyms for entities can be developed in chatbot using configuration and different components available.

Thanks for reading.

Source: Artificial Intelligence on Medium