International Journal of Research in Engineering, Science and Management 26

Volume 4, Issue 4, April 2021

https://www.ijresm.com | ISSN (Online): 2581-5792

Artificial Intelligence Based Farmer Assistant

Chatbot
P. Kaviya1*, M. Bhavyashree2, M. Deepak Krishnan3, M. Sugacini4
1,2,3,4
Department of Information Technology, Sri Venkateswara College of Engineering, Sriperumbudur, India

Abstract: Automatic talk bot will be created. The Automatic talk such a service. By introducing an artificial brain, the web-based
bot will be able to answer user questions without any human bot generates customized user responses, aligned to the desired
assistance. These talk bots will be trained with different type of character. Questions asked to the bot, which is not understood
questions. When user types a question, it will apply Naive Bayes
algorithm to identify appropriate answer from list of trained is further processed using a third-party expert system (an online
question. The talk bot provides answer to the query of the farmer. intelligent research assistant), and the response is archived,
The farmer will have any query; the bot replied the corresponding improving the artificial brain capabilities for future generation
queries by the way of chat and voice. If the farmer has any query of responses.
about the agriculture, the bot will send the answers to them
through the voice or text, If the farmer needs any route assistance, B. Smart Answering Chabot based on OCR and over
the bot will show the route for Google map. In case the farmer generating Transformations and Ranking
forgets agriculture related details or some important queries it will With rapid development of information and communication
remind the farmers.
technology, people are very diverse in education, learning style,
Keywords: Artificial intelligence, K-nearest neighbor, Naive and knowledge improvement methods. This paper presents an
Bayes, Talk bot. approach of converting documents into knowledge of Chatbot
system that enables users to make more benefits of it by asking
1. Introduction and answering questions through the use of electronic
Artificial Intelligence with the help of machine learning documents integrated with simulate system. It is an integrated
could facilitate the farmers with virtual assistants who could system for enrich contents of documents from popular format
help them in real time. They could be able to interact with the such as Portable Document Format (PDF) and digital photos.
farmers in their native language so that they can understand The workflow of this system is started from extracts texts using
them well. This will educate the farmers at any point of time. Optical Character Recognition (OCR) from files, and then
Farmers could be leveraged to take advice and generates questions via over generating Transformations and
recommendations from agriculture pro chatbots. Ranking algorithm, and finally let Chatbot response to the
AI has developed a chatbot which predicts a user query. This user’s question when it is matched with the String pattern.
model uses the generative and retrieval model to predict the user C. Toward the Implementation of a Topic Specific Dialogue
query, other information like soil information, crop fertility. In based Natural Language Chatbot as an Undergraduate Advisor
our project we have applied Artificial Intelligence on data In this work, we explain the design of a chat robot that is
related to agriculture such as soil type, crop, climate, Estimated specifically tailored for providing FAQ Bot system for
profit, Government benefits, Agriculture workshops university students and with the objective of an undergraduate
advisor in student information desk. The chat robot accepts
2. Literature Survey natural language input from users, navigates through the
A. An Intelligent Web-Based Voice Chat Bot Information Repository and responds with student information
This paper presents the design and development of an in natural language. In this paper, we model the Information
intelligent voice recognition chat bot. The paper presents a Repository by a connected graph where the nodes contain
technology demonstrator to verify a proposed framework information and links interrelates the information nodes. The
required to support such a bot (a web service). While a black design semantics includes AIML (Artificial Intelligence
box approach is used, by controlling the communication Markup Language) specification language for authoring the
structure, to and from the web-service, the web-service allows information repository such that chat robot design separates the
all types of clients to communicate to the server from any information repository from the natural language interface
platform. The service provided is accessible through a component. Correspondingly, in the experiment, we
generated interface which allows for seamless XML constructed three experimental systems (a pure dialog systems
processing; whereby the extensibility improves the lifespan of associated with natural language knowledge based entries, a

*Corresponding author: kaviya1373@gmail.com

P. Kaviya et al. International Journal of Research in Engineering, Science and Management, VOL. 4, NO. 4, APRIL 2021 27

domain knowledge systems engineered with information voice to deliver the farmer query to bot. It won’t give any better
content and a hybrid system, combining dialog and domain solution at the time of farmer requirement. It takes much time
knowledge). Consequently, the information repository can to resolve the farmer’s query.
easily be modified and focused on particular topic without
recreating the code design. Experimental parameters and 5. Proposed System
outcome suggest that topic specific dialogue coupled with A farmer assistant bot is built using Naive Bayes algorithm
conversational knowledge yield the maximum dialogue session that analyzes user’s queries and understand user’s message.
than the general conversational dialogue. This System is a web application which provides answer to the
D. Chatbots in Education query of the farmer. Farmer just have to query through the bot
which is used for chatting and voice. Farmer can chat using any
Sofie Roose and ruth lochan (1878) proposed the chatbots in
format there is no specific format the user has to follow. The
education and the survey is based on digitalizing education and
System uses built in Naive Bayes to answer the query. The
reinventing the learning experience through E-learning. The
answers are appropriate what the user queries. The User can
pedagogical agents are used to enhance the learning and
query any farmer related activities through the system. The user
teaching and this virtual agent called hikayat land involves
does not have to personally go to the office for enquiry. The
more provisions which included folktales and moral values. The
System analyzes the question and then answers to the user. The
other part in these chatbot includes the answering of FAQs
system answers to the query as if it is answered by the person.
through AIML it also uses the latent sematic analysis. It all
With the help of Naive Bayes, the system answers the query
emphasis on the keywords which are present and the subject
asked by the farmer. The system replies using an effective
and the course which were available.
Graphical user interface which implies that as if a real person is
E. Chatbots on Industries talking to the user. The user can query about the farmer related
Lavin Agarwal (1898) proposed on a review on chatbot activities through online with the help of this web application.
offering industries and the survey is based on grouping the A talk bot, which is a virtual conversational assistant. Farmers
keywords and ads and through this Way it shows the relevant can interact with the bot in a very simple manner. The focus is
ads which the customer is searching for. And this describes the on developing the bot in a more intellectual way, that it can even
implementation of a chatbot framework with an interface to big understand not so well grammatically defined sentences. Bot
data and also with A.L.I.C.E artificial linguistic internet uses the Natural Language Processing technique to parse the
computer entity. Through these keywords which are negative user queries. This bot is both retrieval and generative based. It
its helpful to avoid people seeing others ads when they search can retrieve responses if the query is already defined in the
for. database or it will fetch responses from the query in bot. To
make the responses more understandable, the responses go
3. Algorithm Used through some analytical process, so that analytical responses
K-nearest neighbor or K-NN algorithm basically creates an could be generated. Bot also have an ability to understand user
imaginary boundary to classify the data. When new data points speech and can generate speech based outputs for user queries.
come in, the algorithm will try to predict that to the nearest of
the boundary line.
1. The k-nearest neighbor algorithm is imported from the
scikit-learn package.
2. Create feature and target variables.
3. Split data into training and test data.
4. Generate a k-NN model using neighbors value.
5. Train or fit the data into the model.
6. Predict the future.
We applied Naive Bayes Machine Learning algorithm to
understand the similar question in trained query. Based on the
identification, answer for appropriate query will be displayed.
Step 1: Separate by Class.
Step 2: Summarize Dataset.
Step 3: Summarize Data by Class.
Step 4: Gaussian Probability Density Function.
Step 5: Class Probabilities Fig. 1. Block diagram

4. Existing System 6. Module Implementation

This system fully manual work. If user have any query based A. Train Chatbot
on farming they need to type the query on chat box. They can We will train the chatbot with hundreds of question and
take their own time to response the user. There is no proper answers.
P. Kaviya et al. International Journal of Research in Engineering, Science and Management, VOL. 4, NO. 4, APRIL 2021 28

B. Mobile App Voice Input The proposed system is fully automated the query will be
To pass the user query through the mobile App. answered basis the question and knowledge base automatically.
hence no need to have a person to answer the queries and ease
C. Speech to Text Conversion for the users to interact with queries. Farmer can interact with
A speech recognition engine (or speech recognizer) takes an the talk bot in a very simple manner. The focus is on developing
audio stream as input and turns it into a text transcription. The the bot in a more intellectual way that it can even understand
talk bot will have abilities to hear what user says and respond not so well grammatically defined sentences. Bot uses the
appropriately by saying it via microphone. For the talk bot to Natural Language Processing technique to parse the user
hear what user says, the speech recognition API will be used. queries.
Computer will receive a voice input which contains a structured
sentence. The voice will be inputted to speech recognizer. 8. Conclusion
Several examples of the recognizer that can be used such as This work introduced the fundamentals of chatbots and it
Google Speech Recognition API. The output of the recognizer gives an overview about ideas, products and platforms both
would be a chunk of text or copula or a JSON object which from the past and the present works. The current interest in
contains the text output. chatbots, its influence in agriculture, its advantages and the
D. Http Request with Text limitations were being discussed here in detail. Different
Web servers handle these requests by returning response aspects of the implementation of a chatbot and its working with
messages that contain the requested resource. conversational interfaces have been presented through the
creation of an exemplary chatbot, which included interaction,
E. Text to Speech the user experience design and a general reusable software
Bot would not be so human without ability to speak. As a architecture of chatbots.
machine that only knows 1 and 0, but can only talks if it is The goal of the project was to give an overview about the
designed to. Text to Speech (TTS) is one of the ways to make chatbots, its welfare for agriculture, and its uses for farmers.
computer talking. The input to the TTS API (Google TTS) This knowledge helps exploring further possibilities of chatbot
would be a reply text from bot API. The text would processed usage in agriculture and it enables farmers to apply chatbots to
by TTS API and the output would be a voice file which contains new scenario and thereby it improves efficient and an effective
an appropriate response to the input from user. way of farming.
F. Agriculture Data
9. Future Works
Collect the farmer queries and train the datasets.
1. Chatbots will be almost indistinguishable from humans.
G. Pre-process Chatbot query According to Gartner, chatbots will power 85 percent of all
We will remove stop words using NLTK corpus. Perform customer’s service interactions by the year 2020. NLP will
stemming to identify similar words. grow more efficient, capable of handling more complex
commands and delivering more poignant outputs. Chatbots
H. Machine Learning Algorithm will also be able to have multi linguistic conversions, not
We will apply Naive Bayes Machine Learning algorithm to only understanding hybrid languages like ‘HINGLISH’ with
understand the similar question in trained query. Based on the NLU but with advanced NLG we’ll also be able to
identification, answer for appropriate query will be displayed. reciprocate.
2. The biggest development in future chatbots will be of
7. Results and Discussion creating ‘sympathy chatbots’. To bet on conversational UIs
over top based products as free of endless navigations,
unbounded experience for the users, on track right from
start, listen and learn are the efficiencies in future.

References
[1] C.I. Nass and S Brave, wired for speech: how voice activates and
advances the human – computer relationship: MIT press Cambridge, vol
6, no.3, pp, 55-23, 2005.
[2] J. Bang, H. Noh, Y. Kim and G.G. Lee, "Example-based chat oriented
dialogue system with personalized long-term memory," 2015
International Conference on Big Data and Smart Computing
(BIGCOMP), Jeju, 2015.
[3] E. Haller and T. Rebedea, "Designing a Chat-bot that Simulates an
Historical Figure," 2013 19th International Conference on Control
Systems and Computer Science, Bucharest, 2013.
[4] S. J. du Perez, M. Lall and S. Sinha, "An intelligent web-based voice chat
bot," EUROCON 2009, EUROCON '09. IEEE, St.- Petersburg, 2009.

Fig. 2. Result
P. Kaviya et al. International Journal of Research in Engineering, Science and Management, VOL. 4, NO. 4, APRIL 2021 29

[5] Y. Chen, W. Wang and Z. Liu, "Keyword-based search and exploration [7] Y. Chen, W. Wang and Z. Liu, "Keyword-based search and exploration
on databases," 2011 IEEE 27th International Conference on Data on databases," 2011 IEEE 27th International Conference on Data
Engineering, Hannover, 2011. Engineering, Hannover, 2011.
[6] V. Bhargava and N. Maheshwari,” an intelligence speech recognition
system for recognition system”, 2009.