This is a part of the CS F407 Artificial Intelligence Course at BITS Pilani Hyderabad Campus. In this assignment, a messenger bot is created which can provide intelligent dialog conversations with the user, similar to ELIZA or ALICE.
- AIML is used for the core of the chatbot.
- The messenger bot uses a knowledge base with multiple random responses exchanged between the user and the bot.
- The bot is able to give replies to multiple interactions and stores the conversations in a database. These persistent copies of the interactions can be retrieved the next time user chats with the bot.
- The bot interacts with web services and helps the user decide the course to be taken based on the recent trends in the industry or the recent trending jobs.
- The bot recommends a subject based on domain of interest or based on the professor in whose class he/she is interested in.
- The bot can tell the current weather by interacting with a web API.
- No machine learning is used and the whole project is based on keyword matching and intent recognition.
A corpus of professors and courses is stored in the MySQL database. Artificial Intelligence Markup Language (AIML) is used for creating the core of the chatbot. The frontend is developed using Android and backend using Python. The back end is in Python and front end is created using Android. The app has been tested locally on Android emulator(in Android Studio) and not on an Android Phone.
The database information is stored in MySQL database on a local machine. For viewing and updating databases, mySQL.txt contains the appropriate code.
A database named user_chats should be created in the mysql system first.2 tables named chats and faculty are used for persistent storage.
The chats table stores the timestamp,userid, bot query and bot response in that order.
The faculty table stores the name of the faculty along with the courses taught.
Before starting ensure that you have mysql on your machine.
- Create a database named user_chats
- Create two tables namely chats & faculty
- Change the username and the password in the python files app & conversation
You can refer mySQL.txt for basic SQL commands.
The backend server has to be run first,on port 8000.
The backend server can be run simply via
$ python app.py (Python v2.7 was used at the time of development)
The MyApplication folder contains the frontend code files.
The front end Android part can be run in Android Studio.
- Set up Android Studio,and install an emulator for testing/running the app.
- Open the MyApplication folder in Android Studio.
- Press the Play button to run the app.
- This code was run on Nexus S API 28 device at the time of development.
