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

net/publication/327667155

Chat-Bot For College Management System Using A.I

Article · September 2018

CITATIONS READS

63 12,058

4 authors, including:

Kumkum Bala Sahil Pandita

Bharati vidyapeeth's college of engineering lavale pune South Indian Education Society's Graduate School of Technology
4 PUBLICATIONS 107 CITATIONS 1 PUBLICATION 63 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Kumkum Bala on 15 September 2018.

The user has requested enhancement of the downloaded file.

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 11 | Nov -2017 www.irjet.net p-ISSN: 2395-0072

Chat-Bot For College Management System Using A.I

Prof.K.Bala1,Mukesh Kumar2, Sayali Hulawale3, Sahil Pandita4

1Professor, Computer Department, Bharati Vidyapeeth’s College of Engineering, Lavale, Pune, Maharashtra
2,3,4 Student, Computer Department, Bharati Vidyapeeth’s College of Engineering, Lavale, Pune, Maharashtra
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract- A chat-bots aims to make a conversation between their queries. The answers will be given using the artificial
both human and machine. The machine has been embedded intelligence algorithms. Users won't have to go personally to
knowledge to identify the sentences and making a decision the college for inquiry.
itself as a response to answer a question. Chat-bots will be
completely based on a text-based user interface, allowing the The Users has to register to the system and has to login to
user to type commands and receive text as well as text to the system. After login user can access the various helping
speech response. Chat-bots are usually stateful services, pages. There will be various helping pages through which the
remembering previous commands in order to provide user can chat by asking queries related to college activities.
functionality. It can be utilized securely by an even larger The system will reply to the user with the help of effective
audience when chat-bots technology is integrated with graphical user interface (GUI). The user can query about the
popular web services. The college inquiry chat-bots will be college-related activities with the help of this web
built using artificial algorithms that analyze user's queries application. College-related activities such as annual day,
and understand user's message. The response principle is sports day, Intake and other cultural activities. It will help
matching the input sentence from a user. The User can ask the the students/user to be updated about the college activities.
question any college-related activities through the chat-bot
without physically available to the college for inquiry. The 2. RELATED WORK
System analyses the question and then answers to the user.
With the help of artificial intelligence, the system answers the [1]Question Answering (QA) systems can be identified as
query asked by the students. The system replies using an information accessing systems which try to answer to
effective Graphical User Interface as if a real person is talking natural language queries by providing answers instead of
to the user. The user just has to register himself to the system providing the simple list of document links. QA system
and has to login to the system. The chat-bots consists of core selects the most appropriate answers by using linguistic
and interface that is accessing the core in (MySQL).Natural features available in natural language techniques. They differ
language processing technologies are used for parsing, mainly from the knowledge sources, the broadness of Dialog
tokenizing, stemming and filtering the content of the Systems (NLDS) is an appropriate and easy way to access
complaint. information. QA system based on Semantic enhancement as
well as the implementation of a domain-oriented based on a
KEYWORDS: NLP (Natural language processing), pattern-matching chat-bots technology developed within an
Sentiment Analysis, synsets, Word Net industrial project (FRASI). The proposed approach simplifies
the chat-bots realization which uses two solutions. First one
1. INTRODUCTION is the ontology, which is exploited in a twofold manner: to
construct answers very actively as a result of an deduction
Chatbot (also known as a talkbot, chatterbox, Bot, IM bot or process about the domain, and to automatically populate,
Artificial Conversational Entity) is a computer program that off-line, the chat-bots KB with sentences that can be derived
mimics human conversations in its natural format including from the ontology, describing properties and relations
text or spoken language using artificial intelligence between concepts involved in the dialogue. Second is to pre-
techniques such as Natural Language Processing (NLP), process of sentences given by the user so that it can be
image and video processing, and audio analysis. reduced to a simpler structure that can be directed to
existing queries of the chat-bots. The aim is to provide useful
Chat-bot for college management system project will be information regarding products of interest supporting
developed using artificial intelligence algorithms that will consumers to get what they want exactly. The choice was to
analyze users queries. This system will be a web application implement a QA system using a pattern-matching chat-bots
which will provide answers to the analyzed queries of the technology.
user. Users will just have to select the category for queries
and then ask the query to the bot that will be used for [2]This paper describes an approach to the idea of
answering it. Artificial intelligence will be used to answer the identifying the most important facts in texts describing the
user's queries. The user will get the appropriate answers to life of a historical figure for building a conversational agent

© 2017, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 2030
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 11 | Nov -2017 www.irjet.net p-ISSN: 2395-0072

that could be used in middle-school CSCL scenarios. This As the complaint description can change from person to
paper presents a method for building a chat-bot that can person. The same question may be asked differently from
simulate a historical figure. The can receive as "input" a plain multiple users. One user ask a question so simply and clearly
text or a web page about the historical figure and has as while another user may ask the same question with more
"output" a trained conversational agent which is able to negatively. So it is necessary to find what is the exact
answer all kind of questions about the life experience of that technical issue with the particular product to give a correct
specific person. the purpose is to provide a generic solution solution
to this problem, so the goal is not to simulate the life and
behavior of

[3]Chat-bots are mainly to used to provide conversation

between both human and machine. Admin feeds some
knowledge to the machine so that machine can identify the
sentences and taking a decision itself as a response to
answer a question. The chat used is actually Indonesian
conversational pattern and the database used in this project
is MySQL. It can miss in defining a sentence and how to the
response it while connecting chat application to the
database. So knowledge representation and implementation
of SQL in the pattern-matching operation are needed. A data
that has been modeled based on the pattern of the
conversation would be tested by the help of a series of
scenarios. The conversation with the chat-bots would be
crosschecked back to the basic pattern. It is done so that it Fig -1: Architecture Diagram For ChatBot
can add some knowledge to the database as it has not been
modeled before. If in case the input sentences in the
3. Answer the Complaints
database did not match then it will be remodeled.
As described above whenever user submits a complaint, the
3. PROPOSED SYSTEM negation level and exact issue/question of the complaint are
detected. Then it is checked that is there such question
1. User Login and Complaint: registered in database. If the answer is found then that
answer is sent to that User. If a particular question is not
User registers himself/herself on Chat-Bot application. Then found in the database such questions are answered by
submits his/her complaints and queries regarding the admin person. Once he answered the question the answer is
electronic and home appliances purchased. sent to that user. And that question along with answer is
stored in database so that whenever such questions will be
2. Chat BOT Responding System: asked so that they get answered directly from the database.
Due to this admin doesn’t need to answer same question
a. NLP Processing and Sentiment Analysis for manually anymore
Complaint:
Interface. WorldNet is a lexical and semantic database for the
When user complaint is submitted to the system, NLP is English language. It is used to group English words into the
applied and sense of the complaint is detected. The sense of set of synonyms called synsets, it provides short definitions
the words is found using part of speech tagging and wordnet and usage examples, and records a number of relations
dictionary.By Using the sentiment analysis negation level of among these synonym sets or their members.
a complaint is detected. And user complaints are prioritized
accordingly. 4. PROPOSED METHOD

b. Search Questions in knowledge database: Algorithm Used:

Once the negation level of the complaint is detected, 1. Porter Stemmer Algorithm
furthermore, the exact question in the complaint is detected
using WorldNet. Porter stemming algorithm (or 'Porter stemmer') is a
process for removing suffixes from words in English.
Removing suffixes automatically is an operation which is

© 2017, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 2031
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 11 | Nov -2017 www.irjet.net p-ISSN: 2395-0072

especially useful in the field of information retrieval. pre-set threshold, the entry of wi in r 1 is filled with the
Following are the steps of this algorithm:- index number of i w ~ in T 1 3. If the above two searches fail,
the entry of wi in r 1 is null. After applying the above
[1]Gets rid of plurals and -ed or -ing suffixes procedure for given sentences T 1 and T 2 , the word order
vectors are r 1 and r 2 respectively. For the example
[2]Turns terminal y to i when there is another vowel in the sentence pair, we have: r 1 = {1 2 3 4 5 6 7 }, r 2 = {1 7 3 4 5 6
stem 2} Thus a word order vector is the basic structural
information carried by a sentence. The task of dealing with
[3]Maps double suffixes to single ones: -ization, -ational, etc. word order is then to measure how similar the word order in
two sentences is. For measuring word order similarity of two
[4]Deals with suffixes, -full, -ness etc. Takes off -ant, -ence, sentences the proposed measure is as follows:
etc. Removes a final –e.

2. Word Order Similarity Between Sentences

Let's consider a particular case to illustrate the importance

of word order. For example, for two sentences: 5.CONCLUSION

T1: A dog jumps over the lazy fox. We create a software tool which will be used by any
T2: A fox jumps over the lazy dog. company to help the users to freely upload their queries.
Once the complaint is registered in the database, automatic
These two sentences containing words are exactly same and tokens are generated and conveyed to the customer through
most words appear in the same order. The only difference is a text message and email for further tracking of the
that dog appears before fox in T 1 and dog appears after fox complaint. Natural language processing technologies are
in T 2. As above given two sentences contain the same used for parsing, tokenizing, stemming and filtering the
words, any methods based on "bag of word" give a decision content of the complaint. The output is fed to the algorithm
that T 1 and T 2 are exactly the same. However it is clear for where the strength of the sentence is calculated. The
a human interpreter that T 1 and T 2 are only similar to intensity of negation is calculated, which helps prioritize the
some extent. T 1 and T 2 are dissimilar only in word order. complaint automatically for the service provider to resolve
Therefore any efficient computational method for sentence the complaint.
similarity must take into account the impact of word order.
Sentences consisting of exactly the same words but in In this way, the proposed system will help many
different orders may result in very different meanings. It is organizations to ensure quality service provision and
easy for humans to process word order information. customer satisfaction with less human efforts.
However the incorporation of order information in to
computational methods for understanding natural language REFERENCES
is a difficult challenge. This may be the reason why most
existing methods do not tackle this type of information. In [1] Agnese Augello, Giovanni Pilato, Alberto Machi' ICAR -
this section we introduce a method that stores the Istituto di Calcolo e Reti ad Alte Prestazioni CNR -
information of word order into account when computing Consiglio Nazionale delle Ricerche Viale delle Scienze ,
sentence similarity. Let's assume that for the given two 978-0-7695-4859-3/12 $26.00 © 2012 IEEE . "An
sentences, the joint word set is T. Recall the above Approach to Enhance Chatbot Semantic Power and
mentioned two sentences T 1 & T 2, their joint word set is: T Maintainability: Experiences within the FRASI Project".
= {A dog jumps over the lazy fox} . A unique index number [2] Emanuela Haller, Traian Rebedea Faculty of Automatic
has been assigned to each word in sentences T1 & T2 Control and Computers university Politehnica of
respectively. The index number is simply the order number Bucharest, 978-0-7695-4980-4/13 $26.00 © 2013 IEEE.
that the word appears in the sentence. For example, the "Designing a Chat-bot that Simulates a Historical
index number is 4 for dog and 6 for over in T 1 . In Figure".
computing word order similarity, a word order vector r is
formed for T 1 and T 2 respectively based on the joint word [3] Bayu Setiaji, Ferry Wahyu Wibowo , Department of
set T . For each word wi in T, we try to find the same or a Informatics Engineering STMIK AMIKOM Yogyakarta,
similar word in T 1 as follows: 1. If T 1 contains an Yogyakarta, Indonesia, 2166-0670/16 $31.00 © 2016
occurrence of the same word, we fill the entry for this word IEEE "Chatbot Using A Knowledge in Database-Human-
in r 1 with the corresponding index number in T 1 . to-Machine Conversation Modeling".
Otherwise we try to find the most similar word i w ~ in T 1 .
2. If the similarity between wi and i w ~ is greater than a

© 2017, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 2032
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 11 | Nov -2017 www.irjet.net p-ISSN: 2395-0072

[4] Dungeon Lee, Kyo-Joong Oh, Ho-Jin Choi School of

Computing, Korea Advanced Institute of Science and
Technology (KAIST), Daejeon, Korea, 978-1-5090-3015-
6/17/$31.00 ©2017 IEEE." The ChatBot Feels You – A
Counseling Service Using Emotional Response
Generation".

[5] Wen Hua, Zhongyuan Wang, Haixun Wang, Member,

IEEE, Kai Zheng_, Member, IEEE, and Xiaofang Zhou,
Senior Member, IEEE." Understand Short Texts by
Harvesting and Analyzing Semantic Knowledge".

[6] Anuja P Jain, Asst. Prof Padma Dandannavar , Computer

Science, and Engineering, Gogte Institute of Technology,
Belgaum, India. 978-1-5090-2399-8/16/$31.00_c 2016
IEEE.

© 2017, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 2033

View publication stats