International Journal On Engineering Technology and Sciences – IJETS

ISSN (P):2349-3968, ISSN (O): 2349-3976 Volume IX - Issue VI, June 2022

College Enquiry Chatbot System using Artificial Intelligence

Mrs.K.Malarvizhi1, Assistant Professor, Department of Computer Science and Engineering, Akshaya
College of Engineering and Technology, Coimbatore,TamilNadu, India

Gayathri.V2, Assistant Professor, Department of Computer Science and Engineering, Akshaya College of
Engineering and Technology, Coimbatore,TamilNadu,
Saranya.V3, Assistant Professor, Department of Computer Science and Engineering, Akshaya College of
Engineering and Technology, Coimbatore,TamilNadu,
Vijetha.A4, Assistant Professor, Department of Computer Science and Engineering, Akshaya College of
Engineering and Technology, Coimbatore,TamilNadu,
Vijey.A5, Assistant Professor, Department of Computer Science and Engineering, Akshaya College of
Engineering and Technology, Coimbatore,TamilNadu,
ABSTRACT A
This project aims to develop a college enquiry Chabot that answers any queries post by students like college
details, course-related questions, locationof the college, fee structure etc. The College Enquiry Chatbot project K
is built using Deep learning algorithms that analyses user’s queries and understand the user's message. This
System is a web application that provides answers to the query. Any individual just has to query through the
bot. The answers are appropriate to what the user queries. The User can query any college-related activities
through the system. The user does not have to personally got the college for enquiry. The System analyses the
question and then answers to the user. The user can also give their suggestions through the suggestion box. The
system replies using an effective Graphical User Interface which implies that as if a real person is talking to
theuser.

Keywords: AI, Chatbot, User, Queries, Message

I. INTRODUCTION
This project aims to develop a college enquiry college-related activities through the system. The
Chabot that answers any queries post by students user does not have to personally got other college
like college details, course-related questions, for enquiry. The System analyses the question and
location of the college, fee structure etc. The then answers to the user. The user can also give
College Enquiry Chatbot project is built using their suggestions through the suggestion box. The
machine learning algorithms that analyze user’s system replies using an effective Graphical User
queries and understand the user's message. This Interface which implies that as if are al person is
System is a web application that provides answers talking tot he user. This project is focusing on
to the query. Any individual just has to query creating a chat botto be used by students to get
through the bot. The answers are appropriate to their queries responded easily from the college
what the user queries. The User can query any website. A chat bot is a

293
 International Journal On Engineering Technology and Sciences – IJETS
ISSN (P):2349-3968, ISSN (O): 2349-3976 Volume IX - Issue VI, June 2022

programwhichcandorealconversationswithtextual botscananswerlimitedsetofquestionsandcannot
and/or auditory methods. Using Artificial perform function outside of the code. The other
Intelligence (AI), chatbots can simulate human categoryischatbotsbasedonAIormachinelearning
conversations.Therearetwocategoriesofchatbots. algorithms,these botscananswer ambiguous
One category is command based chatbots where questions which means the user do not have to be
chatbotsrelyonadatabankofrepliesandheuristics. specific while asking questions. Thus, these bots
The user must be very specific while asking the create replies for the user’s queries using Natural
questions so that the bot can answer. Hence,these Language Processing(NLP).

II. LITERATUREREVIEW

CollegeEnquiryChatbotusingRasaFramework probabilisticmodelwiththehelpofRecurrentNeural
ThegrowthoftechnologieslikeArtificialIntelligence Network(RNN).
(AI), Big Data & Internet of Things (IoT), etc. has
ChatbotSystemforCollageFAQ’sUsingArtificialIntel
marked many advancements in the technological
ligence
worldsincethelastdecade.Thesetechnologieshavea
A chat larva (also called a talk Bot, chatterbox,
wide range of applications. One such application is
Artificial informal Entity) may be a worm that
“Chatterbot or “Chatbot”. Chatbots are conversational
conductsaoralcommunicationviasensorysystemor
AIs, which mimics the human while conversing &
matterways.Suchprogramsareusuallydesignedto
eliminates the need of human by automating
convincinglysimulatehoweveranindividual'swoul
mundane tasks. In the study undertaken, we have
d
createdachatbotineducationdomain&itisnamed
haveasaconversationalpartner,therebypassingthe
as“CollegeEnquiryChatbot”,Thischatbotisaweb-
Turing test. Chat bots are typically employed in
basedapplicationthatanalysesandunderstandsuser's
dialog systems for numerous sensible functions as
queriesandprovidesaninstantandaccurateresponse.
well as client service or information acquisition.
Rasatechnologyisusedtoconstructthischatbot.
Chatbotareoftenintegratedintothedialogsystems
It's anopen- of, for example, automated online
sourcetechnology,whichusesitstwomain assistants,giving them the ability of, for example,
packagesi.e.,RasaCore&RasaNaturalLanguage small talking or engaging in casual conversation
Understanding (NLU) in order to build sun related to the
aContextual AI Chatbot. NLU is used to infer the scopesoftheirprimaryexpertsystems.
intent and to extract the necessary entities from
College Enquiry Chat Bot project will be built
user input & the Rasa Core provides the output by
using artificial intelligence algorithms. That will
building a
analyze users queries and understand user’s
message.This system will be a web application

294
 International Journal On Engineering Technology and Sciences – IJETS
ISSN (P):2349-3968, ISSN (O): 2349-3976 Volume IX - Issue VI, June 2022

which can give himself to the system and must login to the system.
answerstothequeriesofthescholars.Studentscan once login user will access to the various she ping
simplyoughttochoosetheclassforthedepartment pages Various helping pages has the both rough
queriesthenraisethequestiontothelarvawhichwill whichtheusercanchatbyaskingqueriesrelatedto
be used for chatting. computer science collegeactivities.The system replies to the user
areaccustomed answer the scholars queriesstudent. with the help of effective graphical user interface.
Thestudent The user can query about the college related
willgettheappropriateanswerstotheirqueries.The activities. Through on-line with the assistance of
answers will be provide mistreatment the inbuilt this internet application. The user will question
computersciencealgorithms.Studentswonhaveto school connected activities such as date and
traveltothecollegetomaketheenquiry. timing of annual day, sports day, and other
cultural activities. This system helps the to be
The system replies using ineffective Graphical user
updated about the college activities.
interface which suggests that as if a true person is
reprehensiontheuser.Theusersimplymustregister

III. PROPOSEDSYSTEM • AcknowledgeWhenQuestionsareBeyondTheir

TheProposedSystemwasdeeplearningwithtensor
Scope.
flow using deep neural network. A chatbot is
• CustomerExperience Comes First.
often
Deeplearningisactuallyasubsetofmachinelearning. It
describedasoneofthemostadvancedandpromising
expressions of interaction between humans and technically is machine learning and functions in
machines. However, from a technological point of thesamewaybutithasdifferentcapabilities.
view,achatbotonlyrepresentsthenaturalevolution The main difference between deep and machine
of a Question Answering system
learning is, machine learning models become well
leveragingNatural Language Processing (NLP). Our
progressivelybutthemodelstillneedssomeguidance. If
Proposed Methodis Feed Forward Neural Network
(FFN) and Natural a machine learning model returns an inaccurate
LanguageProcessing(NLP)thefeedforwardneural prediction then the programmer needs to fix that
networkwasthefirstandbesttypeofartificialneural problem explicitly but in the case of deep learning,
network devised in thisnetwork. the model does it by him. Automatic car driving

Advantages systemisagoodexampleofdeeplearning.

• Simple User Interface Design and Deep Learning, as a branch of Machine Learning,
User Experience. employs algorithms to process data and imitate the
• PersonalizedConversation. thinking process, or to develop abstractions. Deep
• Ability toLearn. Learning (DL) uses layers of algorithms to process

295
 International Journal On Engineering Technology and Sciences – IJETS
ISSN (P):2349-3968, ISSN (O): 2349-3976 Volume IX - Issue VI, June 2022

data, understand human speech, and visually Part of speech tagging, also called grammatical
recognize objects. Information is passed througheach tagging, is the process of determining the part of
layer,withtheoutputofthepreviouslayerproviding speechofaparticularwordorpieceoftextbasedon
inputforthenextlayer.Thefirstlayerinanetworkis itsuseandcontext.Partofspeechidentifies‘make’as
called the input layer, while the last is called an averbin‘Icanmakeapaperplane,’andasanounin
output layer. All the layers between the two are ‘Whatmakeofcardoyouown?’
referredtoashiddenlayers.Eachlayeristypicallya
Word sense disambiguation is the selection of the
simple, uniform algorithm containing one kind of
meaningofawordwithmultiplemeaningsthrougha
activationfunction.
processofsemanticanalysisthatdeterminetheword
FeatureextractionisanotheraspectofDeepLearning. thatmakesthemostsenseinthegivencontext.For
Featureextractionusesanalgorithmtoautomatically example,wordsensedisambiguationhelpsdistinguish
construct meaningful “features” of the data for the meaning of the verb 'make' in ‘make the
purposes of training, learning, and understanding. grade’ (achieve)vs.‘makeabet’(place).
Normally the Data Scientist, or programmer, is
Named entity recognition, or NEM, identifies
responsible for featureextraction.
words or phrases as useful entities. NEM identifies
Speechrecognition,alsocalledspeech-to-text,isthe ‘Kentucky’ as a location or ‘Fred’ as a man's
task of reliably converting voice data into text name.Co-reference resolution is the task of
data. Speech recognition is required for any identifying if and when two words refer to the
application that follows voice commands or same entity. The
answers spoken questions. What makes speech mostcommonexampleisdeterminingthepersonor
recognition especially challenging is the way objecttowhichacertainpronounrefers(e.g.,‘she’=
people talk quickly,slurring words together, with ‘Mary’),butitcanalsoinvolveidentifyingametaphor
varying emphasis and intonation, in different or an idiom in the text (e.g., an instance in which
accents, and often using incorrectgrammar. 'bear'isn'tananimalbutalargehairyperson).

data into an understandable data set. In other words,

Data Collection whenever the information is gathered from
varioussourcesitiscollectedinrawformatthatisn’t
We are collecting College details, Student details,
possible for theanalysis.
Course details, College facilities, etc., for the dataset
Train and Test Data
Data Preprocessing
• For choosing a model we split our dataset into
Pre-processingreferstothetransformationsapplied train andtest
toourdatabeforeprovidingthedatatothealgorithm. • Heredata’saresplitinto3:1ratiothatmeans
DataPreprocessingtechniqueisusedtoconvertthe raw • Trainingdatahaving70percentandtestingdata

296
 International Journal On Engineering Technology and Sciences – IJETS
ISSN (P):2349-3968, ISSN (O): 2349-3976 Volume IX - Issue VI, June 2022

having 30percent V. REFERENCES

• In this split process preforming based on [1].Ms.Ch.Lavanya Susanna, R.Pratyusha,
train_test_splitmodel P.Swathi,P.RishiKrishna,V.SaiPradee,CollegeEnqu
• AftersplittingwegetxtrainxtestandytrainytestMo iryChatbot,InternationalResearchJournalofEngine
delCreation eringandTechnology(IRJET),e-ISSN:2395-0056,p-
• Contextualise machine learning in ISSN:2395-0072,Volume:07
your organisation.
Issue: 3Mar 2020pp 784-788.
• Explore the data and choose the type of
algorithm. [2].Assistant Prof Ram Manoj Sharma, Chatbot
• Prepareandcleanthedataset. basedCollegeInformationSystem,RESEARCHREVI
• Split the prepared dataset and perform cross EWInternationalJournalofMultidisciplinary,ISSN:2
validation. 455-3085(Online),Volume-04,Issue-03,March-
• Perform machine learningoptimization. 2019,pp109-112
• Deploy themodel. [3].P.Nikhila,G.Jyothi,K.Mounika,Mr.CKishorKumar

Model Prediction Reddy and Dr. B V Ramana Murthy on

College enquiry chat-bot is a simple web based ChatbotsUsingArtificialIntelligence
application which aims to provide information ,InternationalJournalofResearchandDevelopment,
regardingcollege.Thechat-botusesNaturalLanguage Volume VIII, Issue I, January/2019,ISSNNO:2236-
Processing and Artificial Neural Network to have 6124, pp1-12.
conversation with humans. The responses of this
chat-bot are programmed up to some extent which [4].PayalJain,CollegeEnquiryChatBotUsingIterativeMode
trainsthemodelforgivingresponses. l,InternationalJournalofScientific Engineering and
Research (IJSER),ISSN(Online):2347-
IV. CONCLUSION 3878,Volume7Issue1,January2019, pp 80-83.
Thegoalofthesystemistohelpthestudentstostay [5].SagarPawar,OmkarRane,OjasWankhade,Pradnya
updated with their college activities. Artificial Mehta, A Web Based College
Intelligent is the fastest growing technology EnquiryChatbotwithResults,InternationalJournalofI
everywhere in the world, with the help of Artificial nnovative Research in Science, Engineering
Intelligent and Knowledgeable database. We can andTechnology,ISSN(Online):2319-
makethetransformationinthepatternmatchingand 8753,ISSN(Print):2347-
virtualassistance.Thissystemisdevelopingchatbot 6710,Vol.7,Issue4,April2018,pp
basedonandroidsystemsowiththecombinationof
3874-3880.
Artificial Intelligent Knowledgeable database and
virtual assistance. We can develop such chat bot [6].HarshPawar,PranavPrabhu,AjayYadav,VincentMend
which will make a conversion between human and onca,JoyceLemos,CollegeEnquiry Chatbot Using
machineandwillsatisfythequestionraisedbyuser. Knowledge in
Themainmotiveoftheprojectistoreducethework load Database,InternationalJournalforResearchinApplied
on the college's office staff and reduce the Science&EngineeringTechnology(IJRASET),ISSN:23
responsetimetoauser'squery. 21-9653;ICValue:45.98,SJImpact

297
 International Journal On Engineering Technology and Sciences – IJETS
ISSN (P):2349-3968, ISSN (O): 2349-3976 Volume IX - Issue VI, June 2022

Factor:6.887,Volume6,IssueIV,April2018,pp Using A.I, International Research Journalof

Engineering and Technology (IRJET) e-ISSN:2395-
2494- 2496.
0056,p-ISSN:2395-0072,Volume:04Issue:
[7].JincySusanThomas,SeenaThomas,ChatbotUsing
GatedEnd-to- 11| Nov -2017, pp2030-2033.

EndMemoryNetworks,InternationalResearchJourna [10[10].Nitesh Thakur, AkshayHiwrale, Sourabh

lofEngineeringand Technology(IRJET) e-ISSN: Selote,Abhijeet Shinde and Prof. Namrata
2395-0056, p-ISSN:2395- Mahakalkar,ArtificiallyIntelligentChatbot,Universal
0072,Volume:05Issue:03Mar2018, ResearchReports,ISSN:23485612, Volume:04, Issue
:06, July – September2017, pp 43-47.
pp 3730-3735.
[8].Prof. Suprita Das, Prof. Ela Kumar, CitethisArticle

DeterminingAccuracyofChatbotbyapplyingAlgorith Gayathri.V,Saranya.V,Vijetha.A,Vijey.A, Mrs. K.

mDesignandDefinedprocess,4thInternationalConfer Malarvizhi, "College
EnquiryChatbotSystemusingArtificialIntelligence",Int
ence on Computing Communication
ernationalJournalofScientificResearchinComputerScie
andAutomation(ICCCA),2018,978-1-5386-6947- nce,EngineeringandInformationTechnology
1/18/2018IEEE,pp 1-6. (IJSRCSEIT), ISSN : 2456-3307, Volume 8Issue 3,pp.
[9].Prof.K.Bala,MukeshKumar,SayaliHulawale,Sahil 289-293,May-June 2022.
Pandita, Chatbot For College ManagementSystem

298