International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 05 Issue: 03 | Mar-2018 www.irjet.net p-ISSN: 2395-0072

Artificial Intelligence Based Chat-Bot

Sumit Wailthare1, Tushar Gaikwad2, Ketkee Khadse3, Pooja Dubey4
1,2,3B.EStudent, Dept. of Information Technology ,Nagpur Institute of Technology, Nagpur, Maharashtra, India
4Professor, Dept. Of Computer Science &Technology, Nagpur Institute of Technology, Nagpur, Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - A Chat-bot is computer program which conduct a town and want to roam the malls in town he will face many
conversation via auditory or textual method. A Chabot are difficulties. He have to ask many people to guide him to the
software agent that interacts with the user for conversation. mall many of them will take advantage and mislead him as
Chatbot typically serve text based user interface allow input he is new. He will waste his time as well as more money as
from user and receive text as well as auditory from output. he will be misguided. In this project we are using Artificial
Information of chatbot are stored in database which is consist Intelligence to make robot which will help him/her to guide
database who provide by owner of shop and requirement of to the mall. It will help to find ways to the outlets he want to
user. This system will be provided answers to the query of the visit, show the movie s show timings in the mall, show the
user very effectively. User just have to put there requirement to discounts on a particular item in entire mall. This will be
the chatbot which is used for conversation .the system will done in verbal and textual form.
used the AI algorithm to give appropriate answer to the user.
If the answer is invalid then system declares answer is invalid. 1.2 Verbot Engine
This invalid answer can be deleting or modified to by admin.
One of the most popular engine are used for regular expression Currently Verbot only works in Microsoft Windows
base natural language processing engine called verbot. This Verbot is coded almost in C# language and requires
makes it easy designer chat-bot & automates conversation Microsoft .Net 1.1 or higher to execute. The main process
with user. which makes all these happens is verbot4engine.exe
(VERBOT) and agentsvr.exe (Microsoft Agent).
Key Words: User Friendly, Information management
System, Desktop Application, Product Efficiency. Whenever a knowledge base is clicked or added it
will be loaded into the verbot player's memory now when
you type (chat) with verbot your inputs will be compared to
1. INTRODUCTION the inputs in the VKB or CKB files, if the searching founds a
match the programs returns the output from the rule where
1.1 Basic Of Chat Bot that particular input was found, if the input wasn't found
then the engine returns the default outputs from the '*'
Chatbot also known as Chatterbots or chatter robots are (wildcard) provided it was already scripted, if not then
software agents that simulate an entity, usually a human verbot won't say anything now the output string is parsed
counterpart of vague or specifically defined characteristics, with the agentsvr.exe and is converted into voice (if the
with whom the user can interact in a conversation. One of output contains msagent tags it will also parsed by
the first and main goals of Chatbot had always been to agentsvr.exe to reproduce the animations or special
resemble an intelligent human person and make it hard or functions) in the same way advanced features like 'learn',
impossible for the other party of the conversation to 'mem', C# commands are also parsed by the verbot engine.
understand their real nature. With the development of more
and more Chatbots of various architecture and capabilities 2. RELATED WORK
the purposes for their usage has widely broadened. These
Chabot can prove sufficient to fool the user into believing [1] One of the most popular languages for the definition of a
they are talking to a human being, but are very limited in chatbot knowledge base is artificial Intelligence Markup
improving their knowledge base at runtime, and had usually Language (AIML). The interpreter must guarantee the
little to no means of keeping track of all the conversation compliance of properly formed AIML documents, perform all
data. the necessary pre-processing duties for the correct usage of
the chatbot and ensure the correctness of both patterns
Chatbots, which are software agents with an artificial matching of user input and chatbot response. A chatbot is
intelligence that allows them to understand the user input software that is used to interact between a computer and a
and provide a meaningful response according to pre- human in natural language. Naturally, it can extend daily life,
compiled knowledge. The chatbots can be pedagogical such as help desk tools, automatic telephone answering
agents or personifications of historical figures who will be systems, to aid in education, business and e-commerce. In
able to talk about their life and work. A Chat-bot is a general, the aim of chatbot designers is to build tools that
conversational agent that interacts with users using natural help people, facilitate their work, and their interaction with
language. Numerous applications of Chat-bots such as computers using natural language; but not to replace the
Customer Service, call canters etc. If a person is new in the human role totally, or imitate human conversation perfectly.
© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 1060
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 03 | Mar-2018 www.irjet.net p-ISSN: 2395-0072

[2] Presently Chappie is being used as a routing agent rather than just using generalized forms of the most
wherein it can classify the requirement of user into one of common sentence types, I hypothesize the chatbot could
the services provided by business based on the first few generate more interesting conversation.
chats and then transfer it to an agent expert in that service. It
uses natural language processing (NLP) to analyses chats This would still be Case Based Reasoning but rather
and extracts intent of the user with a score similar to the than using generalized data, the program would store past
likes of WIT1. Then it uses this information and AIML conversation explicitly, and mimic a given screen name
(Artificial Intelligence Mark-up Language) to make a personality. The chat-bot would only reply using responses
conversation with the user. Through Chappie, trying to learned from the training corpus, and would thus have more
redefine chat experience in an automated manner. The emotional and personality content than other chatbots.
novelty lies in the way we define our system as not merely a
response generator but an intelligent interface to a response 4. SYSTEM ARCHITECTURE
generator. Then we try to bring counting as a way to avoid
repetitions. Overall Chappie is performing decently, but it
1. Home Screen -The Home page is the starting page
needs more sophisticated algorithms to extract intent and
of the application. It has a Welcome button to be
classify chats more accurately.
click in for entering into the application.
[3] Provides answer to the query of the student very
effectively. Students just have to put their query to the bot
which is used for chatting. The system will use the artificial
intelligence algorithms to give appropriate answers to the
user. If the answer is found invalid, then some system to
declare the answer as invalid can be incorporated. These
invalid answers can be deleted or modified by the admin of
the system. The main objective of the project is to develop an
algorithm that will be used to identify answers related to
user submitted questions. The need is to develop a database
where all the related data will be stored and to develop a Figure 1: Home
web interface. The web interface developed will have two
parts, one for simple users and one for the administrator. 2. Login Activity -This enables the user to login to the
application so as to perform activities. The shop
[4] This paper describes an approach to the idea of owner can login by entering their email and
implementing web-based artificially intelligent chat-bot as a password. If the user enters the wrong credentials,
personal assistant of the user, which stimulates setting and then he/she is not allowed to login. In case the user
initiating meetings of user with his clients. The exchange of is new user for this application then he can register
information happens through email conversations whereas himself by choosing not a User? Click to Sign Up .
its evaluation happens through natural language procession
and natural language generation and AIML files. Using
pattern matching algorithm, a system that can act as a
virtual personal assistant to plan user s work and schedule
his meetings was successfully designed. In terms of the
efficiency of the system to respond within a stipulated time
period, which achieved overall 70% efficiency, it can be
concluded that the system is capable enough to be
implemented in the practical world.

3. PROPOSED SYSTEM

In order to provide a proper guidance to the visitor of a mall

there an virtual robot who guides us with then navigation
and according to the latest discount going on in the shop.
Another advantage concerning the path is solved as we can
search for the shop and get the path. In addition, it will Figure 2: Login
provide us accurate path without wasting our time. The
main objective of the system is to utilize our time and do not 3. Register Activity - This lets the Shop owner to register
waste it. This is practical, reliable and eliminates time loss. A with the application. The shop owner has to register by
further objective is to present a system that can accurately giving a shop name, mobile number, email address, valid
evaluate where we can go for shopping and which path to password and an address. All the shop owner details are
follow. If the chatbot was trained on real conversations, stored in a database. This enables the shop owner to login.

© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 1061
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 03 | Mar-2018 www.irjet.net p-ISSN: 2395-0072

3. CONCLUSIONS

We have surveyed several Chabot systems which succeed in

practical domains like education, information retrieval,
business, e-commerce, as well as for amusement. In the
future, you could imagine chatterbots acting as talking
books for children, chatterbots for foreign language
instruction and teaching chatterbots in general .

REFERENCES

[1] Nikita Hatwar AI BASED CHATBOT International

Journal of Emerging Trends in Engineering and Basic
Sciences (IJEEBS) Volume 3, Issue 2 (March-April 2016).

[2] Pratik Salve College Enquiry Chat Bot International

Journal on Recent and Innovation Trends in Computing
and Communication ISSN: 2321-8169 Volume: 5 Issue:
3.

[3] Aniket Dole Intelligent Chat Bot for Banking System

International Journal for Research in Applied Science &
Figure 3:Registration Engineering Technology (IJRASET) Volume 4 Issue IV,
April 2016.
4. Product Discount Activity -This allows to the shop
owner to add product with discount. [4] Sarthak V. Doshi Artificial Intelligence Chabot in
Android System using Open Source Program-
International Journal of Advanced Research in Computer
and Communication Engineering ISO 3297:2007
Certified Vol. 6, Issue 4, April 2017.

[5] BayuSetiaji Chatbot Using A Knowledge in Database

International Conference on Intelligent System, Modling
and Simulation 2016.

AUTHORS

Sumit Wailthare
Figure 4: Add Product Nagpur Institute Of Technology

5.Search Product Discount Activity- This allows to the

users to search products with discount by product name
or product type.
Tushar Gaikwad
Nagpur Institute Of Technology

Ketkee Khadse
Nagpur Institute Of Technology

Figure 5:search

© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 1062