NLP and ANN Based College Enquiry Chat-bot

Hello Technotizers, this article is focusing on providing some information about a Machine Learning based project which is NLP and ANN Based College Enquiry Chat-bot. in the previous article we had seen a basic chat-bot which had certain limitations. Some of those limitations  have been overcome in this project. Do read the previous article if you have missed it out. AIML Based College Enquiry Chat-bot

Note: This project is only for education purpose and does not aim to promote any institution whatsoever.

1.     Introduction

This project is focusing on creating a simple chat-bot to be used by students to get information about the college without visiting the college website or actual college. A chat-bot is a program which can do real conversations with textual and/or auditory methods. Using Artificial Intelligence (AI), chat-bots can simulate human conversations. There are two categories of chat-bots. One category is command based chat-bots where chat-bots rely on a databank of replies. The user must be very specific while asking the questions so that the bot can answer. Hence, these bots can answer limited set of questions and cannot perform function outside of the code. The other category is chat-bots based on AI or machine learning algorithms, these bots can answer ambiguous questions which means the user do not have to be specific while asking questions. Thus, these bots create replies for the user’s queries using Natural Language Processing (NLP). Making a AI based chat-bot s a tedious job and is difficult to understand the process as compared to a databank based chat-bot. Our project develops a NLP chat-bot as a part of learning to create basic chat-bots. It focuses on giving specific information about the college.

Figure 1: How a Chat-bot Works

Figure 1 shows how a chat-bot works. Whenever a user asks any query, the bot will first analyse the request, then identifies intents and entities, builds a response andsends it back to the user. Now, intents mean intention of the query and entity means details of that query. For example, if a student wants to know the fees of a course then the intent will be fees and entity will be name of the course in this case.

2.     Motivation

Chat-bots are motivated by the need of traditional websites to provide a chat facility where a bot is required to be able to chat with user and solve queries. When live agent can handle only two to three operations at a time, chat-bots can operate without an upper limit which really scales up the operations. Also, if any school or business is receiving lots of queries, having a chat-bot on a website takes off the load from support team. Having a chat-bot clearly improves the response rate compared to human support team. In addition, since millennial prefer live chats over a phone call, they find a chat-bot, which provide a highly interactive marketing platform, very attractive. Furthermore, a chat-bot can automate the repetitive tasks. There can be some scenarios where a business or school receives same queries in a day for many times and support team must respond to each query repetitively. Lastly, the most important advantage of having a chat-bot is that it is available 24/7. No matter what time it is, a user can get a query solved. All these advantages of a chat-bot constitute the motivation to implement a College Enquiry Chat-bot.

Before implementing College Enquiry Chat-bot, various existing chat-bots were reviewed such as Amazon Shopping App, YONO SBI chat-bot, Alexa, etc. In order to understand the requirement of a chat-bot, consider an example of Amazon Shopping App. In this app, when a customer buys an item, he/she does not have any information about how to return the item. To get this information, the customer must call and wait to talk to customer representative for a long time. However, this whole process is tedious for a customer. Hence, Amazon created a chat-bot to answer simple queries of customers. Similarly, the College Enquiry Chat-bot is designed to help students to get information on a fingertips.

3.    Objectives:

• •      With the help of Chat-Bot a student need not have to be physically present in college office for the enquiry and thus gaps the bridge between administration and the students.
• •  Helping students to get basic information about the college.
• •      Creating an easy to use chat-bot for both staff and students which requires user to enter any option provided by the chat-bot.
• •      Students opting for admission in the college can have quick information about the college using this chat-bot.

4.     Limitations of existing system:

• ·             Time consuming and hectic
• ·             Needs personal attendance
• ·             Isn’t available 24/7
• The existing system is not only tedious but also makes a gap between administration and students. Surfing the website for small information can be irritating sometimes. It is rather better to just enter your query and get information on your fingertip. Now-a -days there are many changes that occurred in the Education system with help of advanced technological improvements. Everything is happening over the internet without any difficulty. In previous days for submitting a small application we had to visit that place, but as the days are passing away it’s completely changing. Collecting the applications manually will be hectic procedure and it also needs a lot of manpower. For reducing that manpower and such difficulties many devices or systems were emerged day by day. One of them is a chat-bot which serves as an efficient communication tool between system and users. Chat-bot bridges the gap between traditional enquiry means and users that are particular about time.

5.   Implemented Technique

 

A College enquiry chat-bot project is built using artificial algorithms that analyses user’s queries and understand user’s message. This System is a web application which provides answer to the query of the student. Students just have to query through the bot which is used for chatting. Students can chat just like we do on normal messaging without keeping in mind about case sensitivity. The System uses built in artificial intelligence to answer the query. If the answer found to be invalid, the admin will correct and update the incorrect answer. Admin can view and update the information at any time. System allows admin to delete the invalid answer or to add a specific answer of that equivalent question.

The User can query any college related activities through the system. The user does not have to personally go to the college for enquiry. The System analyses the question and then answers to the user. The system answers to the query as if it is answered by the person. With the help of artificial intelligence, the system answers the query asked by the students. The system replies using an effective Graphical user interface which implies that as if a real person is talking to the user. The user can query about the college related activities through online with the help of this web application. This system helps the student to be updated about the college activities.

6.    Architecture/ Framework

 

System Architecture:

Figure 2: System Architecture

According to the architectural diagram of College Enquiry Chat-Bot System, there are 6 modules which are explained as follows accordingly:

1.  Data (JSON File) – This is the main data file which is programmed in JSON language.

2. Manage – In this module the admin performs the various tasks to fetch into the data various college information requirements like placement sheet, department information, timetable, etc. All this fetched information is then retrieved as a response to the user query accordingly. Admin only has the authority to manage data details.

3. User – This module represents user entering into the system.

4. Chat – In this, User types the query and the bot replies to the user query accordingly. Actual Chatting occurs in this phase only.

5. Upload – In this section admin uploads the common/ general information like, faculty information, fee structures, event and seminar notices, etc. which user may query out during chatting phase. Information that is not available while chatting is uploaded.

6. Exit – This is the phase where user after finishing his work closes the system.

7.  Algorithm and process     design:

Algorithm

Step 1: START: IMPORT MODULES AND CREATE NLP OBJECT

Step 2: CREATE THE APP WITH FLASK

Step 3: IF: mymodel.h5 FILE EXISTS IN OS Load the file.

ELSE:

Run model_train.py

Step 4: DEFINE NECESSARY FUNCTIONS FOR EXTRACTING RESULTS

Step 5: GET TEMPLATE FILE

Step 6: GET REQUEST FROM USER AND STORE IN THE VARIABLE USERTEXT.

IF: RESPONSE IS AVAILABLE IN BRAIN FILE

PRINT response ELSE:

PRINT Alternative statement

Step 7: STOP

Process Design

 

Figure 3: Process Design

1.     User adds a query to the chat-bot through the interface.

2.     The main python program extracts the data from the brain file.

3.     The system matches the query with the intents and checks for availability of response in the data.

4.     If response for the query is available it is extracted and passed on to the chat-bot interface with prints the response for the user.

5.     If response is not available then user is asked to visit the website and information is uploaded in the data by the admin.

6.     Brain  file   learns  information  from   various  JSON   files   that  store information like fees, faculty, academics, infrastructure, etc.

Figure 4: Use case diagram

 

The above diagram shows the functions of the user and the admin.

The user can only chat with the chat-bot, whereas the admin can add information, view information, update information and delete information.

8.     Details of hardware &  software

HARDWARE REQUIREMENTS

• Processor: Intel(R) Core(TM)i3-6100U CPU @2.30Ghz 2.30Ghz Installed.
• RAM: 4.00GB
• System Type: 64-bit operating system,x64-based processor.

SOFTWARE REQUIREMENTS

•      Python 3.8 (nltk, numpy, flask, pickle, tensorflow, spacy)

•      JSON

•      HTML

•      Any browser

9.     Experiment and Results

1.      Running the program

The main python program runs on the command prompt.

This generates an address which indicates the running address of the chat-bot.

2.    Getting the interface

• When the generated address is opened in a browser, we come to the main interface of the chat-bot.
• It is designed for an ease of communication.
• The user is first greeted with a message, after that the user can chat on this interface.

3.      Trying the chat-bot

• ·   Here when we enter Computer HOD we get the name of the HOD and also a link to know more about Computer HOD.

 

• ·  Next, College is entered and we get the college address. Google map link is also provided.

10.   Conclusion and  future work

To conclude, College Enquiry Chat-bot is helpful in guiding students with correct and most up to date sources of information. We can conclude from the above situation that greater the data and more the models and use cases, the better is the reaction produced for the user. It is advantageous for international applicants for queries such as fee payment and academic matters. The system is also not case sensitive which means the user can easily search for query. User can also try a lot of phrases for the same query and still get the correct answer. To improve the current functionalities of College Enquiry Chat-bot, in the future, the scope of the chat-bot can be increased by inserting data for all the departments, training the bot with varied data, testing it on live website, and based on that feedback inserting more training data to the bot. Some of the new features which can be added to the bot are Speech recognition, Integration with multiple channels such as phone call, SMS, Skype, Facebook and Twitter, Integration with services such as password reset and course enrolment and Adding sentiment analysis.

• For source code and detailed information of the project visit NLP and ANN Based College Enquiry Chatbot

With this we come to an end of this article. Hope it was helpful. Do provide your feedback and ideas through comments, it would be highly appreciated. See you soon!

Keep coding and exploring new techs!!