CROP RECOMMENDATION SYSTEM

Jyoti Shekhawat Vempalli Dastagiri

Lecturer Undergraduate
Vivekananda Global University Vivekananda Global University
Jaipur,Rajasthan Jaipur,Rajasthan

T.V Yaswanth Kumar Katam Veera Goutham

Undergraduate Undergraduate
Vivekananda Global University Vivekananda Global University
Jaipur,Rajasthan Jaipur,Rajasthan

ABSTRACT These advanced devices in agriculture enable businesses and

farmers to be additional profitable, efficient, safer, and more
Automating agricultural aspects is a mechanical process environmentally friendly. The rise of digital agriculture and
with or without human intervention in agriculture. Due to its connected technologies has opened a wealth of latest
less space of domestic lands, it has become an important knowledge opportunities. Remote sensors, cameras, and
area of choosing the most suitable crops based on alternative connected devices will gather data twenty-four
prevailing factors in the selected area. In Sri Lankan even hours per day over a complete farm or land. These will
though there are enough knowledge, techniques, and monitor plant health, soil condition, temperature, humidity,
methods which are done manually available in agriculture, etc. the quantity of information these sensors will generate is
there is not any system in which the environmental factors overwhelming. this enables farmers to achieve a far better an
are detected and suggests the user which crop type is best improved understanding of state of matters on the bottom
for farming. This paper is consisting of a theoretical and through advanced technology which will inform them
conceptual platform of Recommendation system through additional regarding their situation more accurately and
integrated models of collecting environmental factors using quickly.
Arduino microcontrollers, Machine learning techniques
such as Naïve Bayes (Multinomial) and Support Vector The environmental data that is gathered by remote sensors
Machine (SVM), Unsupervised machine learning algorithm are processed by algorithms and statistical data which will be
such as K-Means Clustering and also Natural Language understood and helpful to farmers for decision makings and
Processing (Sentiment Analysis) concerned with the keep track of their farms. The more inputs and statistical
Artificial Intelligence to recommend a crop for the selected data collected, and higher the algorithmic rule is at
land with site-specific parameters with high accuracy and predicting the outcomes. And the aim is that farmers will use
efficiency. It has been a major problem to identify what to these technologies to attain their goal of improved harvest by
grow, any man has adequate space in the owner’s land. creating better selections within the field. By implementing
Not only domestic lands but also for farming lands. Why it the system of temperature, soil hydrogen ion concentration
has become a problem is that environmental factors such and soil wetness detection, the information captured are
as temperature, water levels, and soil conditions are processed with an explicit algorithmic rule and passed to a
uncertain as they change from time to time. Due to these centralized database that is connected to different modules of
problems, this solution of crop recommendation system the research, so the main system will predict the most
predicts the user, what crop type would be the most effective crop kind that the farmer should grow to require the
suitable for the selected area by collecting the most outcome of the crop kind that is farmed in a home
environmental factors for plant growth and processing garden or the respectable land area.
them with the trained sub-models of the main of the
system.
General Terms
IOT – Internet of Things 2.LITERATURE SURVEY
Keywords In this literature review of this project, the team sought out
and studied various patents, research papers, documents,
Agriculture, Machine Learning, Arduino, Natural Language and newspapers and magazine articles from various scenes.
processing, Farming The paper states requirements and why they tend to move
into precision agriculture which is due to globalization are
1.INTRODUCTION discussed. Precision agriculture is site-specific farming.
Though Precision agriculture has shown an improvement
Modern farms and agricultural operations ar taken place with time, there exist some issues. As mentioned above site-
more totally different than those many decades agone, specific methods of such systems are needed to be
primarily due to advancements in technology, as well as supervised to get an improved result. Only a few of the
sensors, devices, machines, and knowledge technology. outcomes are provided a particular result. Nevertheless, the
Today’s agriculture habitually uses refined technologies like situation is farming is indispensable since if any default or a
robots, temperature and wetness sensors, aerial pictures, mistake occurs, it might lead to serious damage to
and GPS technology, and lots of complex IOT devices. resources and as well as the plants.

1
In this research, it is proposing a system where the major
factors are taken into consideration at the same time and come
up with a solution so that the system will not be complicated
for the user. As mentioned above in the sentence, the major
factors taken into consideration at once is unlike other models
proposed in previous researchers, this system considers all the
major factors which are essential for plant growth, are
processed together using various algorithms whereas the other
models consider only parameters at once keeping the other
factors constant.
As an example, some tests are carried out to find the rate
evaporation and how the plant growth is affected when there’s
no sufficient water present. And a derived equation is
presented as a result
ETo = K pan × E pan
ETo: reference crop evapotranspiration
Kpan: pan coefficient E pan: pan evaporation Fig 1: Data plot of soil moisture using Linear Algorithm
Even though an equation is proposed, there are some
limitations [4]. Mainly this could be done only for a land with a
less area. This is not suitable for commercialization as the profit
will be low when you use less area for cultivation. The second
limitation is the average rainfall of Sri Lanka mostly suitable
for many crop types to grow without any deficiency. Therefore,
the water level alone itself is not a big issue whereas other
factors are. If there’s enough water for plants but no
temperature that would be a problem as the major
environmental factors have a mutual relationship with each
other in plant growth.
The ultimate goal of the previous researches is also to predict
the best crop type. But once the farmer or user has cultivated
the predicted crop type, the system’s job is done. But in the
system that proposed from this paper, has a feedback system
as well. Even after suggesting the best crop type, the system
can track the plant growth and it provides feedback if the farm
is malnourished. So that the user can take necessary
precautions prior.

3. METHODOLOGY Fig 2: Data plot of Sunlight (lux) using Linear Algorithm

3.1 Dataset Collection
It is needed to have a certain temperature, humidity, soil 3.2Collecting Environment Factors
pH, sunlight, and soil moisture for a plant to be grown
healthy. To receive to good harvest those conditions To compare and predict the initial data set, the
should be satisfied. But those conditions may vary environmental factors needed to be gathered. To collect
according to the plant varieties. The initial data set is environmental factors, Arduino microcontrollers [7] are
collected from the Department of Agriculture Sri Lanka used. Since both the temperature and humidity sensors are
[5], other agriculture books, Agricultural web sites [6], and comprised of a single micro-controller, four sensors are
other reports and research papers. This initial data set was used to collect data. They are sunlight intensity sensor, soil
used to train the crop recommendation model so that the moisture sensor, soil pH sensor, and humidity and
accuracy would be increased. temperature sensor. Those sensors are connected with an
Arduino Wi-Fi module and gathered data is sent to the
Table 1. Sample data set database. The gathered data are cleaned and processed by
using clustering and other algorithms to pass the values to
Crop Temp Hum. pH Sun. Moist. the next component of crop recommending and stored in
the database.
Bean 26 63 7 8 63
Lettuce 23.8 60 6.5 8 60
Carrot 26 98 6.5 8 98
Cabbage 29.4 95 7.1 5 95
Beet 30 99 6.5 6.5 6.5

2
 Fig 3: Arduino Micro-Controllers Connectivity

3.3 Crop Prediction soil condition or any other changes in the selected land. But to
avoid these factors affecting the crop prediction, the farmer’s
Since the environmental conditions differ from region to region, a feedback system is included in the system. Once the crop type
machine learning model is used to predict the best crop type for is recommended, the farmer is asked for details and feedbacks
the selected land. To train the crop recommending model with the regularly through the mobile application to guide the farmer
data collected from the Arduino sensors, machine learning with necessary precautions. To provide the necessary feedback,
algorithms [8] are used to identify the best crop to cultivate with the feedback system is used in the mobile application by
the highest probability of growing. Naïve Bayes & Support vector selecting the crop type. By this, the overall accuracy and the
machine algorithms are used to select the best crop type. reliability of the product are increased with time.
From this model, it decided what type of crops that the farmer
should grow. This is done by analyzing factors of humidity,
temperature, soil moisture, pH level, and sunlight. Mainly the 4.RESULT
system suggests 4 crop types by analyzing the above-mentioned
factors using two machine learning algorithms. Naïve Bayes [9] - The proposed system was successfully implemented in a
Naive Bayes is a technique for constructing classifier models that selected land. The data sampled from the sensors at an interval
assign class labels to problem instances which are represented as of one hour on a normal sunny day is presented in Table I.
vectors of feature values, where the class labels are drawn from Based on the tests conducted and the data collected, it can
some finite set. Support Vector Machine (SVM) [10] - The suggest the best crop to cultivate for maximum harvest. Using
objective of the support vector machine algorithm is to find a the feedbacks collected from the farmer the accuracy of the
hyperplane in N-dimensional space (N — the number of features) predictions is sharpened by neglecting the invalid data. As an
that distinctly classify the data points. example, if the farmers are continuously providing the
negative feedback for cultivating strawberry in Galle, the
system itself learned, and the accuracy of the final output is
increased which means in future it will not suggest strawberry
3.4Monitoring and Feedback to grow in Galle. The farmer can enter his own feedback using
his native language. Since the most Sri Lankans use Sinhala
This proposed system product would mainly identify four types of and English, the custom libraries are implemented, by which
crops according to the environmental factors of the selected plot the system identifies the language significantly by using natural
of land. But the reason to obtain a probability of more than 90% language processing algorithms and the feedback is identified.
for the above-identified crops would be the

3
 According to the prevailing environmental factors in the
selected land, the best suitable four crop types are suggested 6.FUTURE WORKS
to the farmer.
It can do a lot of additional features to the system. Currently, it
takes necessary environmental factors as inputs and suggests a
very suitable crop to be cultivated. But as the next level, the
Automation part can be added as the response system to the
feedback. This can be modified to control the humidity, water
level, etc. according to the need of the farmer.
Presently it takes all environmental factors as inputs to the system,
but as an additional feature, an algorithm can be implemented to
predict the one factor using another two factors. (Example –
predicting the soil pH level from soil moisture and sunlight), so
Fig 4: A sample output given to the farmer that the initial cost of setting up the sensors would be less and can
be easily maintained.
Figure 4 shows a sample result where the farmer gets, once the
environmental factors are entered. PLU code is an attribute that
is used to identify the crop uniquely. The overall accuracy of the
proposed system is more than 92%. As the farmer continue to use 7.REFERENCES
this, the more data system feeds, the more accurate it will be.
Therefore. by the time of long usage, the farmer can obtain more 1. Lakshmi.N, Priya.M, Sahana Shetty, and Manjunath C. R,
the 95% of accuracy from the whole system Crop Recommendation System for Precision Agriculture, vol. 6
Reading, IND: International Journal for Research in Applied
Table 2. Accuracy level of the models in the system Science & Engineering Technology, 2018. [Online] Available:
www.ijraset.com.Ding, W. and Marchionini, G. 1997 A Study
Name of the Module Accuracy on Video Browsing Strategies. Technical Report. University of
Maryland at College Park.
Temperature and Humidity
92%
detection model 2. Remi Schmaltz, “What is Precision Agriculture”, April 2017.
2Sunlight intensityu,soil pH and 95% [Online]. Available: https://agfundernews.com/what-is-precision
soil humidity detection model agriculture.html [Accessed Feb.23, 2020].
Crop recommendation system 90%
3. C. Brouwer and M. Heibloem, Irrigation Water Management:
Monitoring and feedback system 96% Irrigation Water Needs, manual 6 Reading, ITALY: Food and
Agriculture Organization of the United Nations, 1987. [Online]
Available:
http://www.fao.org/3/s2022e/s2022e07.htm#TopOfPage.

4. A. de Carbon, "PRECISION AGRICULTURE: ITS

5.CONCLUSION BENEFITS AND LIMITATIONS", carrhure, 2019 [Online].
In a modern environment with less space and less knowledge of Available: https://www.carrhure.com/precisionagriculture-
agriculture, all the factors are considered from the perspective of benefits-limitations/ [Accessed: Feb.25, 2020]
farmer and plant, and the farmer is properly guided until the
harvesting. Before selecting any plant to grow it is important to 5. Department of Agriculture Sri Lanka (2015), ‘Crop Suitability
have the knowledge and an understanding of the factors that Recommendation for Grama Niladhari Divisions in Sri Lanka’,
affect the cultivation and how to maintain or control them. Natural Resources Management Center, Peradeniya
From this system, these above-mentioned factors are
automatically processed and select the crop type to be 6. Marie ionnotti (2020), ‘Outdoors & Gardning’, The Spruce
cultivated. Web Site
Once the plant is cultivated, the farmer is asked for feedback 7. Rohith Gandhi (2018), ‘Support Vector Machine —
regularly with a time interval of one month. From this feedback Introduction to Machine Learning Algorithms’, Toward Data
taken, the system self-trained, and the accuracy is improvedwith Science Web Site
time and data collected. From this system, the guidance of a
specialist is not needed, and the maintenance is less. Thus,
implementing this system will not have any additional monetary
impact on the user. Since the average land space for a person in
Sri Lanka is less, from the data gathered by sensors it is evident
that this proposed system has above 95% accuracy and suitable
for both rural and urban areas. With the time of the accuracy
would be more increased significantly as the significant amount
of data is collected and the system self-trained by the time of
approximately one year. The experiments conducted over the 4-
6 months proved the reliability and accuracy of the system.
When the whole world is looking forward for new agriculture
consultancy measures, this system will definitely
be a torchbearer.