International Journal of Scientific Research in Engineering and Management (IJSREM)

Volume: 08 Issue: 04 | April - 2024 SJIF Rating: 8.448 ISSN: 2582-3930

ANIMAL DISEASE PREDICTION

Jaypriya Chilampande1, Anushka Chakor2, Samruddhi Patil3, Aditi Bhavsar4,

Prof. Nikhil Deshpande5

Department of Information Technology, Sinhgad College Of Engineering, Pune, Maharashtra, India.

---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The rising concern over animal health and technological innovation. As our understanding of
welfare necessitates the development of efficient and animal health depends, so does our ability to harness
accurate disease prediction methods. The importance of advanced technology to improve diagnostics and
this search lies in its capability to bridge the differences treatment methods. One of the most promising
in disease prediction. By studying data from symptoms, developments in this regard is the application of Machine
and using technologies such as frontend tool, this learning in predicting and diagnosing animal diseases.
framework aims to provide a cost-effective, easily Main aim is to belief that every animal deserves the best
accessible solution, and sustainable solution for all possible care, including of geographical location. We
stakeholders in animal health ecosystem. The research find out the challenges faced by veterinarians and animal
methodology involves the collection and pre-processing healthcare, farmers, and animal care takers in diagnosing
of various datasets including different animals’ species, diseases accurately. There are challenges, this research
diseases, and symptoms. These data have been explores the combination of advanced technologies like
thoroughly analyzed to identify patterns and correlations machine learning with veterinary diagnostics. By
between various health parameters. The initial model is including data-driven approaches, including symptoms,
generated using the machine learning algorithm Random and emerging technologies like machine learning
Forest. The experimental results demonstrate the algorithm like Random Forest, this research aims to
effectiveness of the proposed models in accurately improve animal disease prediction. The significance of
predict animal diseases based on input parameters using this research lies in its potential to bridge the gap in
traditional diagnostic methods. disease prediction, it also offers a cost- effective,
accessible and solution for stakeholders across the animal
Key Words: Animal health, Disease Prediction, health ecosystem. The framework proposed in this
Veterinary diagnostics, Machine learning research involves disease prediction but also provide
early intervention, it leads to economic losses in
INTRODUCTION agriculture, livestock industries and reducing the risk of
Animal health is fundamental to sustainable agriculture, zoonotic disease transmission to humans.
livestock management, and global health security.
Timely identification and innervation in animal diseases
are crucial for maintaining agricultural productivity,
ensuring animal welfare, and safeguarding public health.
Traditional diagnostic methods, reliant on clinical
examination and laboratory testing, have limitations,
especially in remote or underserved areas, early disease
detection and intervention. The field of veterinary
medicine stands at the intersection of expertise and

© 2024, IJSREM | www.ijsrem.com | Page 1

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 08 Issue: 04 | April - 2024 SJIF Rating: 8.448 ISSN: 2582-3930

LITERATURE SURVEY identify biological variations. Information is gathered

and processed through machine learning strategies to
1.Animal Disease Prediction using Machine Learning
pinpoint possible health hazards for the animals. The
Techniques:
findings exhibit the model's effectiveness and accuracy,

Addressing the escalating worries about highlighting that the Support Vector Machine (SVM)

diseases that transmit from animals to humans is achieved an accuracy exceeding 90%. [6]

essential, and employing Machine Learning (ML)

5. A Machine Learning and Optimization Framework
methodologies for detecting and forecasting
for the Early Diagnosis of Bovine Respiratory
illnesses in animals is a prospective strategy. ML
Disease:
can be significantly influential in the early
identification and surveillance of such diseases.[1] In this paper, the authors present a framework for
BRD detection, its early detection, and
2.Application of Random Forest Algorithm on
sustainable BRD scenario using virtual IoT
Feature Subset Selection and Classification and
technologies We use Cost Optimization Worth
Regression:
(COW), devices study including the problem of

Datasets have lots of variables, picking out the costly choices. Let’s take the example. COW

important ones is crucial. Random forest is a powerful increases the accuracy of the forecast due to

tool for this job, even with a ton of variables. It's great economic constraints. We show that COW is NP-

at dealing with missing data, sorting things into Hard, and propose an efficient estimator with

categories, and predicting values. Plus, it can handle polynomial complexity called the Cost Aware

messy data like outliers and noise really well. In our Learning Feature (CALF).We validate our

study, we used random forest to pick out the best methodology on a real dataset collected from 159

features and compare how well it works for different calves during the prehearing period. Results show

tasks like sorting data and making predictions.[8] that our approach outperforms a recent state-of-
the- art solution.[4]
3.Machine learning techniques in predicting
lumpy skin disease occurrence based on 6.SEIR-SEI-EnKF: A New Model for

meteorological and geospatial features: Estimating and Forecasting Dengue Outbreak

Dynamics:
This paper explores the use of ML algorithms to
forecast the occurrence of Lumpy Skin Disease Virus Dengue fever is a severe mosquito-borne
(LSDV) infection in cattle based on meteorological viral disease that causes a significant social

and geological attributes. The research reveals that burden in many tropical and subtropical regions.

certain ML techniques can predict LSDV occurrence Accurate predictions of dengue outbreaks enable
with high accuracy, particularly the ANN (Artificial local health authorities to take proactive measures

neural networks) algorithm, which outperformed such as installing mosquito control equipment or

other methods.[7] repairing medical supplies. We developed a new

model for dengue outbreak estimation and
4.Animal Health Monitoring using Machine prediction: the vector-borne disease model SEIR-
Learning: SEI with Kalman filtering in groups of
susceptible, exposed, infected, recovered
The research suggests a way to use machine
buildings (ma host) and susceptible, infectious,
learning to monitor animals to track their vital signs and
and virulent (for infection). ( EnKF) Method of

© 2024, IJSREM | www.ijsrem.com | Page 2

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 08 Issue: 04 | April - 2024 SJIF Rating: 8.448 ISSN: 2582-3930

assimilation. The SEIR-SIR-EnKF model was Machine Learning Algorithm:

first validated using simulated dengue outbreaks
In this system we use Random Forest Machine
in two studies.[5]
learning algorithm. It is popular learning method used
in machine learning for both classification and
DESIGN AND METHODOLOGY regression tasks. It works by constructing multiple
decision trees during training and outputting the class
1.1 System Architecture:
that is mode of the classes(classification) or the mean
A system architecture is the model that used to
prediction (regression) for individual trees. Random
defines the structure, behavior, and more about
forest is a powerful algorithm known for its flexibility
of system. A system architecture can consist of
and ability to make reliable predictions for a wide range
system components and sub-systems developed,
of data types and functions.
that will work together to Implement the overall
system. Classification:

Classification in machine learning is a

supervised learning task with the goal of classifying input
data into specific classes or groups based on its
characteristic

1.Training Data: A labeled dataset used to train the

classification model, consisting of input features.
The structure of the system refers to
2.Model Training: Random Forest algorithm used for
diagnosis of diseases in animals. Each input data
classification.
record will use. Here is quick explanation of the
steps: 3.Model Evaluation: The trained model’s performance is
evaluated using metrices like accuracy, precision, recall
Input data:
on test dataset.
The machine requires input records. This need data
4. Prediction: Once trained, the model can predict the
which consists of animal type and five different symptoms.
class labels based on learned patterns from training data.
Based on input data which taken from uploaded dataset are
divided into training data and testing data. Training Dataset:The training dataset is a collection of
data which is used to training the ML model.
Preprocessing Techniques:

Result:
In data preprocessing technique the process of
preparing the raw data and making it ready for a machine The final step in the process where system displays the
learning model. It is first important step in machine predicted disease based on the analysis.
learning for creating model. In preprocessing techniques
in ML, primary task is to prepare and clean the raw data 1.2Methodology

before it is used to train a model. This process involves

The partially developed projects described illustrate
tasks like handling missing values, splitting data into
a systematic approach based on symptom-based disease
training and testing sets. The goal is to ensure that the
prognosis. Initially, the dataset involves collecting various
data is in suitable format.
qualitative data from multiple sources, which are then

© 2024, IJSREM | www.ijsrem.com | Page 3

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 08 Issue: 04 | April - 2024 SJIF Rating: 8.448 ISSN: 2582-3930

combined into a centralized repository to facilitate analysis diseases based on the symptoms entered. Then user will be
Data preprocessing is done to cleanse and transform the directed to an output page displaying the predicted disease along
raw data to ensure its suitability for further analysis. with relevant information and with some suggestion of nearest
Symptom extraction follows, using machine learning to doctors. Also, on home page there is feature of chatbot is given,
extract features from symptoms, which are then combined where user can clear their doubt about animal related questions.
to produce more effective disease predictions. Machine In that chatbot the fix number of question are included which are
learning algorithm random forest implemented and trained helpful for user to clear their doubt. The instant responses and
on historical symptoms data to develop predictive models. improving customer service and user experience. Ultimately, the
Model selection considers data characteristics and diseases project highlights how technology can significantly contribute to
to choose appropriate models, which are incorporated into improving animal health outcomes by enabling early disease
the prediction process. A user interface is designed to detection. The outcome of this project is a tool that can detect
facilitate data entry and user interaction. Symptom diseases early by looking at symptoms. The result of this work
information and results are securely stored in a database. is a tool that can detect diseases early by looking at symptoms.
Finally, performance monitoring and feedback This can help farmers, veterinarians and other users take
mechanisms to evaluate system accuracy and user immediate action to treat sick animals and prevent further spread
satisfaction based on actual results and user feedback Such of disease. It’s like an early warning system for animal health
a system this approach ensures a robust and sensitive issues, allowing early intervention to keep animals healthy.
disease prediction system.

RESULT
The result of this research project involves the
development of predictive model for animal diseases
using machine learning, specifically the Random Forest
algorithm. By training this model with data on various
animal symptoms, the system can accurately predict
diseases based on input symptoms. The implementation
includes a simple user interface, likely a website, where
users can input symptoms and receive disease predictions
in an interactive manner. This user-friendly approach
ensures that farmers, veterinarians, and other users can
easily access and benefit from the disease prediction tool.
These results are shown in simple UI.
We have created the simple website to show out output
in a very interactive way i.e.
o Sign Up
o Login
o Main Page
o Output Page

By taking input of symptoms from user the websites backend

will process the input data using machine learning algorithm
which is Random Forest. The model will predict potential

© 2024, IJSREM | www.ijsrem.com | Page 4

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 08 Issue: 04 | April - 2024 SJIF Rating: 8.448 ISSN: 2582-3930

REFERENCES
[1] Sana Rehman.” Animal Disease Prediction
using Machine Learning Techniques,” “IJRASET”,
2023.

[2] Abishaik Mohan, R. Deepak Raju, Dr. P.

Janarthanan. “Animal Disease Diagnosis Expert
System using Convolutional Neural Networks,”
CONCLUSION
“International Conference on Intelligent Sustainable
Animal disease prediction is one of the most important Systems (ICISS 2019) IEEE”, 2019.
areas where technology, data science and veterinary
medicine improve overall animal health and productivity. [3] YU QU, HUAXIANG ZHANG, CHENG
We developed predictive models for animal disease by LIANG, AND XIAO DONG, “KATZMDA: Prediction
using machine learning techniques specifically Random of miRNA-Disease Associations Based on KATZ
Forest and implemented a user-friendly interface for Model,” “IEEE” 2018.
displaying predictions. Our work not only helps animals
stay healthy; it also helps farmers keep their animals [4] ENRICO CASELLA, MELISSA C. CANTOR, JOAO
thriving. From Ahed’s perspective, using tools like this H. C. COSTA, “A Machine Learning and Optimization
can make a huge difference in protecting animal health Framework for the Early Diagnosis of Bovine
and keeping our food safe. The result of this work is a Respiratory Disease,” “IEEE “2023.
tool that can detect diseases early by looking at
symptoms. This can help farmers, veterinarians and other [5] Chunlin Yi, Lee W. Cohnstaedt, And Caterina M.
users take immediate action to treat sick animals and Scoglio, “SEIR-SEI-EnKF: A New Model for
prevent further spread of disease. It’s like an early Estimating and Forecasting Dengue Outbreak
warning system for animal health issues, allowing early Dynamics,” “IEEE” 2021.
intervention to keep animals healthy.
[6] Karthick B,“Animal Health Monitoring using Machine
Learning ” ,“IJCRT”, 2021.
ACKNOWLEDGMENT

We express our profound gratitude to principal of our [7] Ehsanallah Afshari Safavi, “Machine learning
college Prof. S.D. Lokhande and to the Head of the techniques in predicting lumpy skin disease occurrence
department Information Technology Dr. S.R. Ganorkar based on meteorological and geospatial features,”
for allowing us to proceed with project and giving us full “Springer,”14 January 2022.
freedom to access the lab facilities. Our heart full thanks
to our guide Prof. N.H. Deshpande for taking time and [8] Jitendra Kumar Jaiswal, Rita Samikannu , “
helping us throughout our project. And, thanks to the staff Application of Random Forest Algorithm on Feature
member of IT Department, Sinhgad college of Subset Selection and Classification and Regression” ,
engineering. We also appreciate to Kaggle.com for “World Congress on Computing and Communication
providing the datasets used in project. And last but not Technologies (WCCCT)” , 2021.
the least, a hearty thanks to all our classmates for
supporting us throughout the project research.

© 2024, IJSREM | www.ijsrem.com | Page 5

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 08 Issue: 04 | April - 2024 SJIF Rating: 8.448 ISSN: 2582-3930

[9] Alexandra L’Heureux, Hany F. Elyamany , “Machine

Learning with Big data : Challenges and Approaches” ,
“IEEE” , 2017.

[10] A. S. More, Dipti P. Rana, “Review of random forest

classification techniques to resolve data”, “IEEE”,
2017.

© 2024, IJSREM | www.ijsrem.com | Page 6