See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/341592140

Design and Implementation of Student Chat Bot using AIML and LSA

Article  in  International Journal of Photoenergy · April 2019

CITATIONS READS

17 3,231

1 author:

Mohammed Ismail
Kannur University
53 PUBLICATIONS   359 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

fractal compression on cuckoo search and spiral architecture View project

Machine Learning View project

All content following this page was uploaded by Mohammed Ismail on 23 May 2020.

The user has requested enhancement of the downloaded file.

 International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-6, April 2019

Design and Implementation of Student Chat Bot

using AIML and LSA
K.Naga Lakshmi, Y. Kishore Reddy, M. Kireeti, T.Swathi, Mohammad Ismail


Abstract: Now-a-days students are facing many problems going on in the class? " and the chatbot may convey the
regarding the information of a college student details. There is no given by the engineer and uses that information to
proper communication channel to know the required details of the improves the precision after some time by understanding
students in a college. This paper focuses on the process of the obscure discussions.
communication automation on web using computer
programming. By using computer program a conversational
The point of a chatbot is to lead discussion which permits
agent is created which responds to the user statements called chat clients can access to data through a lightweight
bot. It can take the input from user in many formats like text and application by informing. Essentially chatbots are
speech. Using AIML (Artificial Intelligence Markup Language) partitioned into two sorts dependent on the information
and LSA (Latent Semantic Analysis) the relevant answer to the and methods for learning and they are as per the
user query is generated. If the relevant answer to the user query is
following:
not found the system will automatically ping to the admin.
 Rule-based chatbot:
Keywords: AIML: Aritifcial Intelligence Markup Language,
The Standard based chatbot fill in as an intuitive chatbot
LSA: Latent Semantic Analysis
that gives the appropriate response relying upon the
I. INTRODUCTION principles given by the designer. They are additionally
called as the FAQ bot. These sort of chatbots are given a
The idea of creating a machine that had human thought lot of information with the two inquiries and separate
process made many scientists, philosophers and even answers by the software engineer. At the point when the
sculptors fascinated of humanistic automation. Student client gives an inquiry it checks in guidelines given by
Information Chatbot’s are the communication interfaces that the software engineer and reacts in like manner.
help the user to know the details of the students. Chatbot’s  AI chatbot:
came a long back in second half of the 19th century with a AI(Artificial Insight) chatbots go about as fake human
simple conversations and now turned into powerful tool for brain. It utilizes NLP (normal language ) handling
e-commerce and media brands. strategies and as a modern intellectual. It comprehends
The chatbots are broadly classified into two types as follows the two solicitations and setting in the demand alongside
 First one the purpose of the question and feeling of the client. It
It has a lot of predefined rules, conditions, streams and consequently learns the substance structure the client
breakdown conditions pursued by triggers to react to the discussions..
particular inquiries or directions. One straightforward There is much more to learn by the AI chatbots. Now a days
model is the chatbot that tells the understudy data in a we are seeing big market players (players intends chatbots)
specific montage. One individual or client can solicit any like Google Home, Siri and Alex.
number from precedents one such be name of the There are many obstacles that are to be overcome in the
understudy with id number then the visit bot will scan for creation of chatbots are as follows:
the appropriate response and returns the yield. This sort  The bots ought to learn by it self so they ought to have
of chatbots are possibly shrewd when the designer (one the capacity to learn in the intelligent mode while
who builds up the chatbot) is savvy and structure the utilizing by the client. One of the main trouble is
chatbot by deduction each inquiry of the discussion. maintaining the alluring mode for the client and
 Second one makes them to utilize the bot longer time which help
It utilizes the new advances like ML(machine learning) the bot to learn.
and simulated intelligence (Man-made reasoning) that  The chatbots more often than not learn structure the
comprehends the significance of the words for all dialects tenets given by the engineer and the directions given
and comprehends the feeling which don't depend on just by the client. They don't have any ethics to recognize
pre-characterized information. A client can ask "what's which is great and which isn't great.
 A few directions are difficult to instruct to a machine.
Revised Manuscript Received on April 15, 2019. Give us a chance to consider a model understanding
K. Naga Lakshmi, Department of Computer Science & Engineering, the feeling and learning it. This is the hardest piece
KL University, Vijayawada (A.P), India.
of the chatbot challenge contrasted with the
Y. Kishore Reddy, Department of Computer Science & Engineering,
KL University, Vijayawada (A.P), India. remaining however it is the main key for growing
M. Kireeti, Department of Computer Science & Engineering, KL best chatbot.
University, Vijayawada (A.P), India.
T. Swathi, Department of Computer Science & Engineering, KL
University, Vijayawada (A.P), India.
Dr. Md Ismail, Department of Computer Science & Engineering,
Professor, KL University, Vijayawada (A.P), India.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F5079048619/19©BEIESP 1742 & Sciences Publication
 Design and Implementation of Student Chat Bot using AIML and LSA

The first chatbot Eliza which is created in 1996 by the which go about as the base for all later chatbots. Google
individual of MIT ( Massachusetts Establishment of propelled Google Now in 2012 which is cell phone based
Innovation) which mimics the psychotherapist, They are application. Presently it turned into the piece of updates and
developing the lives step by step which has being utilized UI alterations for portable pursuit. Later on in 2015 bots
from most recent two years. One of the hallucination that ended up well known with further developed highlights like
there is best straightforwardness toward the front and there voice collaboration utilizing language preparing calculations
are numerous difficulties to be tackled that make a consistent in Alexa bot and Cortana bot that is available in various
ordeal by the client like improvement of stream, checking dialects which perceives common voice directions. These
blunders API coordination and making deception live human bots likewise performs web seeking. In 2016 Facebook
discussion by acceleration. propelled bots for flag-bearer stage where creates can make
Consider chatbots human-to-robot texting. Utilizing chatbots and collaborate with Facebook clients.
man-made brainpower and cautiously composed PC contents, There are distinctive chatbots available for clients which
chatbots can perceive regular language to have simple made them to utilize them for various purposes, for example,
discussions with your clients – responding to questions and NIKA(for timesheets),ACEBOT(for costs), TWYLA bot (for
associating similarly as another human may. client administration), QnA bot (for FAQ ) and WIZU bot (for
The main reasons for growing of chatbots usage are as inputs).
follows
 Growing Customer Expectations II. LITERATURE SURVEY
 Changing Conditions for Organizations Artificial Intelligence Chat bot in Android System using
 Technological Factors Open Source Program-O:
 The result: Increasing use of chatbots The framework works in two modes, content and voice. At the
A. Reason for Sometimes filing of chatbot: point when client gives the contribution to content
configuration the principal mode is actuated. The client input
Balanced discussion is simple when contrasted with one-with
is passed to the middleware API for the reaction. On other
numerous discussions. Presently chatbots has
hand when client gives the voice input at that point second
one-to-numerous discussions where one server attempt to
mode is initiated, In this voice mode we first proselyte the
speak with different clients. In this the main undertaking is the
voice into content before sending it to middleware API.
manner by which every single discussion is scripted and the
Middleware is the model which interfaces the AIML contents
connection ability with the client. A few times chatbot fail to
with our android application. At the point when client input is
convey the right answer in light of the fact that the direction,
gotten at the middleware, it is passed to the example
inquiry is new to the bot or the failure of the database, because
coordinating calculation which keeps running over the AIML
of the wasteful information , over-burden on the server,
contents. In this procedure, right off the bat the example
obscure feeling utilized by the client.
coordinating calculation is executed for coordinating of the
B. History of chatbots substantial reaction from the available AIML contents. At the
The first chatbot was created by Joseph Weizenbaum in 1966 point when design is coordinated, the comparing layout is
and named as Eliza chatbot which appeared as a come back to the middleware. At that point Middleware
psychotherapist who reacted to the client with essential encodes the layout into the JSON design and sends the answer
inquiries. It gave a deception of comprehension with scripted to the android application. Subsequent to accepting the
reactions. In 1972 Kenneth Colby turned out with a reaction application interpret the JSON and gives the reaction
progressed chatbot Repel than Eliza which is a chatbot that to the client.
could mimic an individual with distrustful schizophrenia. The response generation process is carried out with two
Later on in 1988 an endeavor is made in making chatbots with phases:
AI (Man-made consciousness) through human cooperation A. Preparation of Pattern Matching
and named the chatbot as Jabberwacky. Later on in 1992 B. Pattern Matching Behavior
Imaginative labs for MS-Dos makes a chatbot named Implementation:
Dr.Sbaitso which is consolidated with AI and it is intended to 1. Detailed implementation of text based chatbot.
exhibit a digitized voice. In 1995 utilizing regular language 2. Detailed of the implementation of voice based chatbot..
preparing Counterfeit Phonetic Web PC Element (A.L.I.C.E A. An E-business Chatbot using AIML and LSA:
chatbot) chatbot is create where heuristic example is utilized
For e-plans of action Web based business is one of the for the
for coordinating tenets to human contribution to request to
most part used to work together over the web or web. One of
have discussion. In 2001 a canny bot is made which is
the real downside in this field is client administration quality
generally disseminated crosswise over sms systems called
which they are giving. In every single e-plan of action, clients
smarterchild. Later on IBM's Watson was planned in 2006
need to wait for the reaction for longer time from the client
where normal language handling and machine learning is
administration agent or structure the partner side answer.
utilized to uncover inside and out information from
As an answer for this issue, another chatbot which can
substantial measure of information. In 2010 an astute
consequently offers
individual right hand chatbot named siri appeared which
reactions to the clients with
utilizes common language UI to respond to questions and
in a matter of seconds
perform different demands and is a piece of Apple's IOS

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F5079048619/19©BEIESP 1743 & Sciences Publication
 International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-6, April 2019

(division of seconds) in view of the informational collection III. PROPOSED MODEL

given by the designer and the Every now and again Addressed
The framework works in two modes content and voice. At
Questions(FAQs), utilizing AIML and LSA (Man-made
the point when client gives the contribution to content
reasoning Markup Language and Dormant Semantic
organization the principal mode is actuated. The client input
Examination ).
is passed to the middleware API for the reaction. On other
B. Web Navigation Prediction Using Multiple hand when client gives the voice input at that point second
Evidence Combination and Domain Knowledge: mode is actuated, in this voice mode we first believer the
Fake neural networks(ANN's) system is utilized in this. This voice into content before sending it to middleware API.
paper mainly focuses on the web in which the client Middleware is the model which associates the AIML
movement can be utilized as the information for contents with our web application. At the point when client
recommending the following site pages that the client is going input is gotten at the middleware, it is passed to the example
to open. It utilizes Counterfeit neural systems and markov coordinating calculation which keeps running over the
models. At the point when the client asks for the AIML contents. In this procedure, right off the bat the
WWW(world wide web) the demand is handled and relying example coordinating calculation is executed for
upon the past history of the client the following site page that coordinating of the legitimate reaction from the available
the client may open can be recognized so as to stack it for AIML contents. At the point when design is coordinated, the
quick access of the client. comparing format is come back to the middleware. On the
off chance that the example isn't coordinated, at that point
C. College Enquiry Chatbot Using A.L.I.C.E: the info keeps running over the LSA scripts. If the example
A.L.I.C.E is a characteristic language handling and man-made is discovered the it returns the output to the middleware.
reasoning visit robot.The programming used to make is from
open source (which can be uninhibitedly utilized by
everybody) called Alice bot programming and AIML
programming. Here a chatbot is tailored as an application
which encourages new understudies to take care of few of the
issue they are confronting and responds to inquiries for the
understudies that are raised amid confirmation. AIML is a
mediator that takes the contribution from the client and
applies design coordinating calculation utilizing AIML
contents and creates the ouput. It additionally helps while
infusing the recently formed information or learning into the
information base.
D. Leveraging Question Answer technology to
address terrorism inquiry:
This paper researches the potential utilization of discourse
based ALICE bots in scattering fear mongering data to the
general public.Created the Psychological warfare Movement
Asset Application (TARA) framework, which depends on an
adjusted adaptation of ALICE Program D that can be
Figure 1 Block Diagram show an overview of the
unreservedly obtained from www.alicebot.org. Both TARA
proposed model
and ALICE share a considerable lot of similar segments.
In Middleware encodes the layout into the JSON design and
E. Android based Chat-Bot: sends the answer to the client. In the wake of accepting the
This framework has been executed to incorporate with any reaction web disentangle the JSON and gives the reaction to
Inn The board Android Application to facilitate the procedure the client. On the off chance that the example isn't
of lodging booking. ALICE utilizes design coordinating and coordinated utilizing both AIML and LSA contents it pings
spares the data in Man-made reasoning Increase Language to the administrators telephone as a message. The figure1
(AIML) records. ELIZA is a visit bot that is utilized mainly to demonstrates the diagram of our proposed model.
chat with the client. It is mainly utilized for easygoing
A. Pattern matching:
correspondence with user.chat-bot has worked in information
base, conversational capacities, constant help, and human It is the demonstration of checking a given grouping of
visual look. It can go about as an extraordinary remote helper tokens for the nearness of the constituents of some example.
for lodging booking and can perform multi-dimensional In qualification to design acknowledgment, the match
assessment. commonly must be definite: "it is possible that it will or will
n't be a match." The normally for the most part have the state
of either arrangements or tree structures.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F5079048619/19©BEIESP 1744 & Sciences Publication
 Design and Implementation of Student Chat Bot using AIML and LSA

Employments of example coordinating grasp yielding the has been snared in a repercussion of techniques. for instance,
areas (assuming any) of an example inside a token succession, its rankings cover the ones of people on in vogue vocabulary
to yield some piece of the coordinated example, and to and issue matter appraisals, it copies human expression
substitute the coordinating example with some another token arranging and classification decisions, reenacts
arrangement. word-expression and section word lexical preparing data and,
as expressed in organization Papers, effectively gauges entry
B. AIML:
soundness, learnability of entries by means of individual
Artificial Intelligence Mark-up Language (AIML) is derived understudies and the fine and amount of mastery contained in
from Extensible Mark-up Language (XML) which is used to an article.
build up a conversational agent (chatbot) artificially. In this
we use “program-O‟ which is an AIML interpreter for
generation of the responses of users input. We have used this
method for developing an android application chatbot.
which will interact with user using text and voice responses.

LSA works in the following way:

IV. RESULTS

Among s Natural Language Processing begins when the client

submits an inquiry to Visit bot. This inquiry is gotten by the
AIML mediator that answers utilizing a learning base (Jindal
et al, 2004) and there is a detailed task at Figure 4. This
information base was executed in AIML language comprising
of labels on class structure. In every class we have the inquiry
and the appropriate response, utilizing individually, the
example and layout labels. To improve the collaboration, V. FUTURE SCOPE & CONCLUSION
making it as reasonable as could be expected under the
circumstances, we can utilize more than one response to one In our project we implemented conversation of voice input to
inquiry when we have a similar inquiry, done in various ways. text input and applying AIML scripts for both the inputs and
In the event that the inquiry isn't in the learning base chatbot, a generating the output using which acts as a simple chatbot.
reaction is produced haphazardly by the framework, keeping We can use any interface that acts as a intermediate channel
the client dependably find a similar solution. We can all the between AIML, SQL, and backend. In our implementation we
more likely comprehend the utilization of these labels by are using python as an interface as python in platform
above figure. independent and it contains many inbuild functions and has
many advantages. In upcoming days this chatbot is applied
C. LSA: with LSA and returns the appropriate ouput to the user query
Latent Semantic analysis (LSA) is a thought and approach for and if the ouput is not found by applying both LSA and AIML
separating and speaking to the relevant utilization which the query is pinged to the admin using python as an interface
methods for words by method for factual calculations and android development. This prototype can give upto 90%
actualized to an enormous corpus of literary substance. The accuracy and helps the user to get the response in shorter time
fundamental idea is that the totality of certainties pretty much and with appropriate results. This bot implementation can be
the majority of the word settings wherein a given word does used in different business sectors by utilizing this technique
and does not appear bears a rigid of shared constraints that in and applying for that
extensive part decides the likeness of importance of particular business.
expressions and set of words to each extraordinary. The
ampleness of LSA's reflected picture of human information

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F5079048619/19©BEIESP 1745 & Sciences Publication
 International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-6, April 2019

REFERENCES
1. “Artificial Intelligence Chatbot in Android System using Open Source
2. Program-O” by Sarthak V. Doshi, Suprabha B. Pawar, Akshay G.
Shelar, Shraddha S. Kulkarni published in International Journal of
Advanced Research in Computer and Communication Engineering in
2017
3. “Android based Chat-Bot” by Shabina Sayed, Rushabh Jain published
in International Journal of Computer Applications in 2016
4. Leveraging Question Answer technology to address terrorism inquiry
by Robert P. Schumaker, Hsinchun Chen published in Elesvier in 2007
5. “Web Navigation Prediction Using Multiple Evidence Combination
and Domain Knowledge” by Mamoun A. Awad and Latifur R. Khan in
IEEE
6. “College Enquiry Chatbot Using A.L.I.C.E” by Balbir Singh Bani,
Ajay Pratap Singh published in International Journal of New
Technology and Research (IJNTR) in 2017
7. “An E-business Chatbot using AIML and LSA” by Thomas N T in
presented in Conference on Advances in Computing, Communications
and Informatics (ICACCI) in 2016

AUTHOR PROFILE

K.Naga Lakshmi, is a student at the

department of Computer Science and
Engineering at K L Educational foundation,
Deemed to be University, Vaddeswaram,
Andhra Pradesh. She is doing her research
work in Knowledge Engineering (KE).

Y.Kishore Reddy is a student at the

department of Computer Science and
Engineering at K L Educational foundation,
Deemed to be University, Vaddeswaram,
Andhra Pradesh. he is doing his research
work in Knowledge Engineering (KE).

M.Kireeti is a student at th department

of Computer Science and Engineering at
K L Educational foundation, Deemed to
be University, Vaddeswaram, Andhra
Pradesh. he is doing his research work in
Knowledge Engineering (KE).

T.Swathi, is a student at the department of

Computer Science and Engineering at K L
Educational foundation, Deemed to be
University, Vaddeswaram, Andhra
Pradesh. She is doing her research work in
Knowledge Engineering (KE).

Dr. Md Ismail is a professor at the

department of computer science and
engineering at K L Educational
foundation , Deemed to be University,
Vaddeswaram, Andhra Pradesh.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F5079048619/19©BEIESP 1746 & Sciences Publication
View publication stats