Technical jargon related to IT support service might scare a naive user off, often he may find himself in a situation where he is unable to find the right question that caters to his problems; and also he does not want to chat with a customer service executive: this "pseudo-chat bot" will help you with your needs.
What it does
The user will enter his query as a natural language query into the portal. Based on the keywords present in the query, the software will decide which FAQ best matches the question asked by the user and present it along with a suitable response.
How we built it
We have defined a knowledge base that contains a list of important keyword categorizations. We have two FAQ files which are parsed by our system to create a series of keywords. The user then enters a question at the command line that also gets parsed into keywords. These keywords are matched together and only the relevant answers are displayed.
Challenges we ran into
The knowledge base has to be correct and as complete as possible to ensure a broad range of queries can be covered. Also, defining the rules based on the correct set of keywords and mapping them to the correct question is also a crucial step in the process.
Accomplishments that we're proud of
Ability of the system to communicate with the user with minimal executive intervention. Based on its knowledge base, it is able to correctly predict the correct question and the corresponding response.
What we learned
sASP is a powerful system for doing logical programming and hence the logical part of our system has been coded in sASP.
What's next for IT Support System
We need to handle tenses, etc. in other words, different forms of the same keyword. We also need to account for misspellings on the part of the user.