GROUP 1
ITB Presentation
Presented To: Date:
Dr. Vinay Singh 16-08-2023
Our Team
Kuldeep Mangla (2020IMG-031)
Mahendra Sahu (2023PMSF-101)
Aayush Adoni (2020IMG-001)
Mayank Singh (2020IMG-038)
Smart Crop Advisor System
Based on Soil Nutrition &
Environmental Conditions
Outline 1. Motivation & Problem Statement
Table of Contents 2. What it is?
3. How it works?
4. Proposed Methodology
5. Results
6. Conclusion
7. Novelty & Future Scope
8. References
06
Motivation & Problem Statement
Our country India is an agriculture dominant country.
Our 54.6 % of the population is engaged in agriculture and its related
activities and the share of agriculture in the GDP of our country is about
19.9 % which is close to about 400 billion dollars.
Also, India is the 2nd largest producer of agriculture products in the
world and accounts for 7 percent of global agriculture output.
These figures are really impressive. But here comes the shocking fact:-
Look at this country wise graph showing area of Arable and Permanent Cropland[2]:-
INDIA
CHINA
07
The above graph shows that India has more land on which agriculture is done but
unfortunately China has about twice the agriculture output that of India. China's total
agriculture output is close to 900 billion dollars making[9] it number 1 in the world.
(Remember the India's figures discussed above)
Now a common question arises in our mind, why this is happening?
> China employed two main changes[1] in their agriculture due to which their
production shoots up:-
1. Chemicals & Fertilizers
2. Technology Farming
Chemicals and fertilizers are frequent in India now a days but the key difference is due to
Technology farming. Indian farmers are still using traditional agriculture methods which
leads to a less production. Here, comes my project into limelight which will help farmers to
multiply their production by selecting right crop on right time.
08
Smart Crop Advisor System
What it is?
Smart AI Based Crop Advisor is a modern farming application based on AI and
machine learning in which the system recommends the right crop for farming
based on certain parameters[5] like quantity of different nutrients in the soil,
humidity condition, rainfall level, temperature and PH of soil. This system results
in the selection of the best crop for sowing. As a result, the production will
increase.
10
How
it
works?
The farmer has to fed
values of different
parameters in the
system and it will
show the best crop
for the farm.
LIVE DEMO
Website Click Here
Proposed Methodology
Supervised learning
1. Supervised learning[3] is a type of machine learning where the algorithm learns
from labeled training data to make predictions or decisions.
2. In this approach, the algorithm is provided with input data along with their
corresponding correct output labels.
3. The goal of supervised learning is for the algorithm to learn the underlying
patterns and relationships in the data so that it can generalize and make accurate
predictions on new, unseen data.
11
Regression Classification
In regression,[4] the goal is to predict a continuous In classification, the goal is to assign input data points
numerical value based on input features. The to predefined classes or categories based on their
relationship between the input features and the target features. This involves finding a decision boundary that
variable is modeled as a mathematical function. separates different classes in the feature space.
12
13
Framework
1. Import Libraries
2. Set Up Environment: Configure the environment settings.
3. Define Functions: Define several functions to perform specific
tasks.
4. Load Dataset: Load a dataset from a CSV file.
5. Remove Outliers: Calculate the first and third quartiles (Q1
and Q3) of the data. Calculate the interquartile range (IQR).
Filter out rows where any column value is below (Q1 - 1.5 *
IQR) or above (Q3 + 1.5 * IQR).
6. Split Data: Define the target column name.
7. Train Model:
• Create a pipeline.
• Fit the pipeline to the training data.
• Make predictions on the testing data.
8. Save Model: Save the trained model using ‘pickle‘ to a
specified filename.
14
Process Flow Diagram
13
Results
I have used various Machine Learning models in my project. I am explaining the two
prominent models here:-
KNN
Random Forest Classifier
The dataset I took consists of 11000[8] entries among which the 80% of the data is
used for training and rest 20% data is used for testing. I considered 7 parameters and
included 22 types of crops.
I am getting different accuracies through the different models used.[10] So, here is a
brief look on the output of the program through the two models :-
15
Random Forest Classifier KNN
16
17
Random Forest Classifier KNN
Conclusion
This project draws inspiration towards adopting advanced techniques in Agriculture based
on Artificial Intelligence and Machine Learning.
It will lead to the development of the Agri industry in India and also increase the Gross
Domestic Production (GDP) [7]contribution of agriculture.
This project is essential as our country India is an Agriculture Dominant nation where more
than half of the entire population is engaged in agriculture and its related activities.
The other reason that dedicated me to developing this project is farmers’[11] current poverty
level among farmers, which leads to their high suicide rates. This is very unfortunate for our
nation.
To conclude, there has been an extensive space for research in this field.
18
This system can be used by farmers
Novelty on their mobile anytime anywhere
& without the help of any crop doctor.
We can add other[10] parameters as
Future Scope well such as location and mandi
prices of different crops such that the
farmer can also select the maximum
profiting crop instead of selecting the
maximum production crop.
In future this[8] system will be
beneficial to the whole nation and the
share of agriculture in GDP will also
increase.
20
References
1. “Cunningham, p´adraig and cord, matthieu and delany, sarah jane,”machine learning techniques for
multimedia: Case studies on organization and retrieval”,year=”2008.” [Online]. Available:
https://link.springer.com/chapter/10.1007/978-3-540-75171-7 2# citeas
2. “Cropland area by country, 2017.” [Online]. Available: https://www.worldometers.info/ food-
agriculture/cropland-by-country/
3. “Supervised machine learning: All you need to know.” [Online]. Available: https:
//www.simplilearn.com/tutorials/machine-learning-tutorial/supervised-machine-learning
4. “Regression vs. classification in machine learning.” [Online]. Available: https:
//www.javatpoint.com/regression-vs-classification-in-machine-learning
5. R. K. Rajak, A. Pawar, M. Pendke, P. Shinde, S. Rathod, and A. Devare, “Crop recommendation system to
maximize crop yield using machine learning technique,” International Research Journal of Engineering and
Technology, vol. 4, no. 12, pp. 950–953, 2017. [Online]. Available: https://shorturl.at/fiJW3
6. “Crop recommendation system using machine learning for digital farming, usa.” [Online]. Available:
https://www.prolim.com/ crop-recommendation-system-using-machine-learning-for-digital-farming/
5.
6. 21
7. “Nidhi h kulkarni; g n srinivasan; b m sagar; n k cauvery, ”improving crop productivity through a crop
recommendation system using ensembling technique,” 2019.” [Online]. Available:
https://ieeexplore.ieee.org/document/8768790
8. “Farmeasy: Crop recommendation for farmers made easy, to recommend optimum crops to be cultivated
by farmers based on several parameters, 2020.” [Online]. Available: https://
towardsdatascience.com/farmeasy-crop-recommendation-portal-for-farmers-48a8809b421c
9. “Soumya sri attaluri, nowshath k batcha, raheem mafas, ”crop plantation recommendation using feature
extraction and machine learning techniques,” 2020.” [Online]. Available:
https://www.researchgate.net/publication/345247916 Crop Plantation Recommendation using Feature
Extraction and Machine Learning Techniques
10. “Crop recommendation dataset, maximize agricultural yield by recommending appropriate crops, 2020.”
[Online]. Available: https://www.kaggle.com/datasets/atharvaingle/ crop-recommendation-dataset
11. “Dr. a. k. mariappan, ms. c. madhumitha, ms. p. nishitha, ms. s. nivedhitha,”crop recommendation system
through soil analysis using classification in machine learning”, 2020.” [Online]. Available:
http://sersc.org/journals/index.php/IJAST/article/view/30399
