Claims:WE CLAIMS

1) Our Invention Intelligent Multiple Regression Analysis in Agriculture: Crop Productivity, Rainfall, Fertility, Humidity, Temperature Using Machine Learning is a Machine learning support tool for crop yield prediction, including supporting decisions on what crops to grow and what to do during the growing season of the crops. The invented methods and process combined analysis of variance was done from the mean data obtained for each characteristic over two seasons and correlation and regression analysis were carried out to better understand the relationship between yield and some yield components. The Results indicated that seasons significantly affected all traits and interaction between seasons and cultivars was also significant and also Highly significant differences and adequate genetic variability were observed among cultivars for all the eight characters. The invented results also of the correlation coefficients of traits with grain yield revealed that the grain number per spike (r=0.84**), grain weight/spike (0.87**), 1000-grain weight (r=0.88**), number of spikes per square meter (r=0.68*) and spike length (r=0.67*) had the highest significant positive correlation with grain yield, indicating dependency of these characters on each other. The Best Subset Multiple Regression analysis indicates that adding the variable number of grain per spike (X4) and grain weight per spike (X5) does not improve the fit of the model. The invented technology is a Crop yield may be assessed and predicted using a piecewise linear regression method with break point and various weather and agricultural parameters, such as NDVI, surface parameters (soil moisture and surface temperature) and rainfall data and the parameters may help aid in estimating and predicting crop conditions.
2) According to claim1# the invention is to a Multiple Regression Analysis in Agriculture: Crop Productivity, Rainfall, Fertility, Humidity, Temperature Using Machine Learning is a Machine learning support tool for crop yield prediction, including supporting decisions on what crops to grow and what to do during the growing season of the crops.
3) According to claim1,2# the invention is to a methods and process combined analysis of variance was done from the mean data obtained for each characteristic over two seasons and correlation and regression analysis were carried out to better understand the relationship between yield and some yield components.
4) According to claim1,2,3# the invention is to a Results indicated that seasons significantly affected all traits and interaction between seasons and cultivars was also significant and also Highly significant differences and adequate genetic variability were observed among cultivars for all the eight characters.
5) According to claim1,2,4# the invention is to a results also of the correlation coefficients of traits with grain yield revealed that the grain number per spike (r=0.84**), grain weight/spike (0.87**), 1000-grain weight (r=0.88**), number of spikes per square meter (r=0.68*) and spike length (r=0.67*) had the highest significant positive correlation with grain yield, indicating dependency of these characters on each other.
6) According to claim1,2,5# the invention is to a Subset Multiple Regression analysis indicates that adding the variable number of grain per spike (X4) and grain weight per spike (X5) does not improve the fit of the model.
7) According to claim1,2,4,6# the invention is to a technology is a Crop yield may be assessed and predicted using a piecewise linear regression method with break point and various weather and agricultural parameters, such as NDVI, surface parameters (soil moisture and surface temperature) and rainfall data and the parameters may help aid in estimating and predicting crop conditions.
Data: 10/5/21
Deepak Sitaram Sonawane
Dr. Ashok Yeshwantrao Tayade
Mr. Devdatta Kashinath Mokashi
Sagar Parashuram Dhamone
Dr. Nilesh Mahajan
Mrs. Sangeeta Patil
Lalita Kiran Wani
Anuradha Jape
Dr. Yogita Vishal Bhapkar
Jyoti Atul Dhanke
Suvarna Ranjeet Jagtap

, Description:FIELD OF THE INVENTION
Our Invention is related to an Intelligent Multiple Regression Analysis in Agriculture: Crop Productivity, Rainfall, Fertility, Humidity, Temperature Using Machine Learning.
FORM 2
THE PATENT ACT 1970 &
The Patents Rules, 2003
COMPLETE SPECIFICATION
(See section 10 and rule 13)
TITLE OF THE INVENTION:
Intelligent Multiple Regression Analysis in Agriculture: Crop Productivity, Rainfall, Fertility, Humidity, Temperature Using Machine Learning.

FIELD OF THE INVENTION
Our Invention is related to an Intelligent Multiple Regression Analysis in Agriculture: Crop Productivity, Rainfall, Fertility, Humidity, Temperature Using Machine Learning.
BACKGROUND OF THE INVENTION
Machine learning (ML) approaches are used in many fields, ranging from supermarkets to evaluate the behavior of customers (Ayodele, 2010) to the prediction of customers’ phone use (Witten et al., 2016). Machine learning is also being used in agriculture for several years (McQueen et al., 1995). Crop yield prediction is one of the challenging problems in precision agriculture, and many models have been proposed and validated so far.
This problem requires the use of several datasets since crop yield depends on many different factors such as climate, weather, soil, use of fertilizer, and seed variety (Xu et al., 2019). This indicates that crop yield prediction is not a trivial task; instead, it consists of several complicated steps. Nowadays, crop yield prediction models can estimate the actual yield reasonably, but a better performance in yield prediction is still desirable (Filippi et al., 2019a). Machine learning, which is a branch of Artificial Intelligence (AI) focusing on learning, is a practical approach that can provide better yield prediction based on several features. Machine learning (ML) can determine patterns and correlations and discover knowledge from datasets.
The models need to be trained using datasets, where the outcomes are represented based on past experience. The predictive model is built using several features, and as such, parameters of the models are determined using historical data during the training phase. For the testing phase, part of the historical data that has not been used for training is used for the performance evaluation purpose. An ML model can be descriptive or predictive, depending on the research problem and research questions. While descriptive models are used to gain knowledge from the collected data and explain what has happened, predictive models are used to make predictions in the future (Alpaydin, 2010).
ML studies consist of different challenges when aiming to build a high-performance predictive model. It is crucial to select the right algorithms to solve the problem at hand, and in addition, the algorithms and the underlying platforms need to be capable of handling the volume of data. To get an overview of what has been done on the application of ML in crop yield prediction, we performed a systematic literature review (SLR).

A Systematic Literature Review (SLR) shows the potential gaps in research on a particular area of problem and guides both practitioners and researchers who wish to do a new research study on that problem area. By following a methodology in SLR, all relevant studies are accessed from electronic databases, synthesized, and presented to respond to research questions defined in the study. An SLR study leads to new perspectives and helps new researchers in the field to understand the state-of-the-art.
Barley (Hordeum vulgare L.) is one of the most important cereal crops in Egypt and is ancient as the origin of agriculture itself. It is considered as one of the most suitable cereal crops, which can survive and grow over a wide range of soils and under many adverse climatic conditions compared with many other cereal crops. It ranks fourth after wheat, rice and maize in the world's cereal production. In most of the crop improvement programmes, raising grain yield is one of the major objectives.
The information on association between grain yield with its components is prerequisite for breeding programmes aiming at yield improvement. Therefore, association and regression studies were undertaken in barley. In recent years, breeding of new barley varieties with high yield and good quality has been regarded as a very important research approach by agricultural science researchers in our country, and several new varieties of barley with high yield and good quality have been developed.
Yield in barley is a very complex trait and is a result of the interaction between various yield components. Knowledge of the association between yield-related traits is of immense importance to the selection of desired combinations of characters. Further, correlation analysis provides information about the correlated responses to selection of important plant characters. Correlation and regression analyses are multivariate tools that help to study the interrelationships and inter-dependence among traits.
In many crops, especially cereal crops, yield depends on some plant attributes such as plant height, number of leaves, stalk thickness and tillering capacity etc. These plant attributes are referred to as the World Essays J. Vol., 1 (3), 88-100, 2013 89 independent variables, covariates, predictors, or regressors, while yield is the corresponding dependent variable. Each of these regressors contributes to the variation in the yield of a variety, although the contribution varies from one variety to another. Correlation analysis among yield and yield components is one

of the prerequisite techniques to determine the influence of environment on productivity and yield potential.
The information on the nature and magnitude of correlation coefficients help breeders to determine the selection criteria for simultaneous improvement of various characters along with yield. Determination of correlation coefficients between various barley characters helps to obtain best combinations of attributes for obtaining higher return per unit area. The statistical technique that is used to establish the existence of linear relationship between the dependent variable and the independent variables is the Regression Analysis.
If there is a single independent or predicator variable is referred to as simple linear regression, while if it involved more than one independent or predictor variables we have the case of Multivariate regression or multiple regression analysis.
The aim of Multiple Regression Analysis (MRA) is to find the best set of the independent variables which can explain dependent variable on condition that the assumptions are provided. The term regression is used to establishing the actual relationship between two or more variables. But scientific, social, economic and agricultural phenomena do not confine to two variables. In these studies, we often need to give actual relationship between two or more than two variables (Agrawal 1991). For this purpose, we choose the method of all possible regressions. This technique requires that investigator fit all the subset regression models involving one predictor variable, two predictor variables and so on. Each subset regression model was then evaluated according some suitable criterion like R2, R2 -adjusted and Mallow's Co.
Monitoring crop conditions is important for the economic development of any nation, particularly developing ones. The use of remote sensing has proved to be very important in monitoring the growth of agricultural crops and irrigation scheduling. Efforts have been made to develop various indices for different crops for different regions worldwide. Generally, crop production has a direct impact on year-to-year variations on national and international economies and food supply.
Weather and climatic conditions are important parameters in controlling the yield of agricultural crops. Crop-weather relations have been used for predicting crop yield. Least-square regression is a principle method usually applied into crop-weather relationships, which may be complex and non-linear. However, conventional approaches of forecasting crop yield using ground based data collection is tedious, time consuming and often difficult.

Using remote sensing data, efforts have been made to develop various indices such as vegetation condition index (VCI), thermal condition index (TCI) and normalized difference vegetation index (NDVI). VCI and TCI are well known for drought detection, monitoring excessive soil wetness, assessment of weather impacts on vegetation and evaluation of vegetation health and productivity. NDVI reflects vegetation greenness; it indicates levels of healthiness in the vegetation development.
Though vegetation development of crop fields may differ from those of natural vegetation because of human influences involved such as irrigation, fertilization and pesticidal activity, NDVI can be a valuable source of information for the crop conditions. NDVI data may be used extensively in vegetation monitoring, crop yield assessment and forecasting.
OBJECTIVES OF THE INVENTION
1. The objective of the invention is to a Multiple Regression Analysis in Agriculture: Crop Productivity, Rainfall, Fertility, Humidity, Temperature Using Machine Learning is a Machine learning support tool for crop yield prediction, including supporting decisions on what crops to grow and what to do during the growing season of the crops.
2. The other objective of the invention is to a methods and process combined analysis of variance was done from the mean data obtained for each characteristic over two seasons and correlation and regression analysis were carried out to better understand the relationship between yield and some yield components.
3. The other objective of the invention is to a Results indicated that seasons significantly affected all traits and interaction between seasons and cultivars was also significant and also Highly significant differences and adequate genetic variability were observed among cultivars for all the eight characters.
4. The other objective of the invention is to a results also of the correlation coefficients of traits with grain yield revealed that the grain number per spike (r=0.84**), grain weight/spike (0.87**), 1000-grain weight (r=0.88**), number of spikes per square meter (r=0.68*) and spike length (r=0.67*) had the highest significant positive correlation with grain yield, indicating dependency of these characters on each other.
5. The other objective of the invention is to a Subset Multiple Regression analysis indicates that adding the variable number of grain per spike (X4) and grain weight per spike (X5) does not improve the fit of the model.
6. The other objective of the invention is to a technology is a Crop yield may be assessed and predicted using a piecewise linear regression method with

break point and various weather and agricultural parameters, such as NDVI, surface parameters (soil moisture and surface temperature) and rainfall data and the parameters may help aid in estimating and predicting crop conditions.
SUMMARY OF THE INVENTION
Crop yield prediction is an essential task for the decision-makers at national and regional levels (e.g., the EU level) for rapid decision-making. An accurate crop yield prediction model can help farmers to decide on what to grow and when to grow. There are different approaches to crop yield prediction. This review article has investigated what has been done on the use of machine learning in crop yield prediction in the literature.
During our analysis of the retrieved publications, one of the exclusion criteria is that the publication is a survey or traditional review paper. Those excluded publications are, in fact, related work and are discussed in this section. Chlingaryan and Sukkarieh performed a review study on nitrogen status estimation using machine learning (Chlingaryan et al., 2018). The paper concludes that quick developments in sensing technologies and ML techniques will result in cost-effective solutions in the agricultural sector.
Elavarasan et al. performed a survey of publications on machine learning models associated with crop yield prediction based on climatic parameters. The paper advises looking broad to find more parameters that account for crop yield (Elavarasan et al., 2018). Liakos et al. (2018) published a review paper on the application of machine learning in the agricultural sector. The analysis was performed with publications focusing on crop management, livestock management, water management, and soil management. Li, Lecourt, and Bishop performed a review study on determining the ripeness of fruits to decide the optimal harvest time and yield prediction (Li et al., 2018).
Mayuri and Priya addressed the challenges and methodologies that are encountered in the field of image processing and machine learning in the agricultural sector and especially in the detection of diseases (Mayuri and Priya, xxxx). Somvanshi and Mishra presented several machine learning approaches and their application in plant biology (Somvanshi and Mishra, 2015). Gandhi and Armstrong published a review paper on the application of data mining in the agricultural sector in general, dealing with decision making.

They concluded that further research needs to be done to see how the implementation of data mining into complex agricultural datasets could be realized (Gandhi and Armstrong, 2016). Beulah performed a survey on the various data mining techniques that are used for crop yield prediction and concluded that the crop yield prediction could be solved by employing data mining techniques (Beulah, 2019). According to our survey of review articles, the significant ones of which are presented in this section, this paper is the first SLR that focuses on the application of machine learning in the crop yield prediction problem.
The existing survey studies did not systematically review the literature, and most of them reviewed studies on a specific aspect of crop yield prediction. Also, we presented 30 deep learning-based studies in this article and discussed which deep learning algorithms have been used in these studies.
The present invention teaches a method to predict crop yield. It also teaches a program storage device readable by a machine that may store a machine readable language to predict crop yield. Furthermore, it is also a crop yield predicting device.
In one aspect of the invention, crop yield may be predicted by receiving data, such as vegetation factors, identifying an initial piecewise linear empirical equation with at least one break point that may be applicable to data received, and determining coefficients of such equation by performing an optimization process.
Any crop from any geographical location may be examined. Data received with respect to a particular crop may be based upon a select increment of time according to a range of time. When predicting its yield, it is preferable to use data averages from the selected increment of time.
The optimization process may perform multiple iterative convergences on the identified initial piecewise linear empirical equation with at least one break point to optimize coefficients. Additionally, this process may use a non-linear Quasi-Newton method. This method may apply a loss function.
Additional objects, advantages and novel features of the invention will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.
Science direct:

The search string is [“machine learning” AND “yield prediction”] (Title, abstract, keywords) and [((“machine learning” OR “artificial intelligence”) AND “data mining” AND (“yield prediction” OR “yield forecasting” OR “yield estimation”))(Title, abstract, keywords).
Scopus:
The search string is [“machine learning” AND “yield prediction”](Title, abstract, keywords) and [((“machine learning” OR “artificial intelligence”) AND “data mining” AND (“yield prediction” OR “yield forecasting” OR “yield estimation”))] (Title, abstract, keywords).
Web of Science:
The search string is [“machine learning” AND “yield prediction”] (title, abstract, author keywords, and Keywords Plus). Springer Link: The search string is [“machine learning” AND “yield prediction”](anywhere) and [((“machine learning” OR “artificial intelligence”) AND “data mining” AND (“yield prediction” OR “yield forecasting” OR “yield estimation”))] (anywhere)
Wiley:
The search string is [“machine learning” AND “yield prediction”] (anywhere).
Google Scholar:
The search string is [“machine learning” AND “yield prediction”] (anywhere) and [((“machine learning” OR “artificial intelligence”) AND “data mining” AND (“yield prediction” OR “yield forecasting” OR “yield estimation”))] (anywhere).
For Web of Science and Wiley, the search string [((“machine learning” OR “artificial intelligence”) AND “data mining” AND (“yield prediction” OR “yield forecasting” OR “yield estimation”))] did not result in any publications.
Cultivation Practices
Sowing date was Dec. 3 and 5 in the first and second season, respectively. Sowing was done by hand in plots of 5 rows, 3 m long and 20 cm wide with plants spaced 5 cm apart within rows. Sowing rate was 60 kg seed/feddan for all genotypes. Grains were drilled in rows using dry method of planting. The preceding crop was maize in both seasons.
Fertilizers were applied at the rate of 100 kg /fed ammonium nitrate (33.5% N) in two equal doses, the first dose was added at tillering stage and the second dose was added at shooting stage, while phosphorus and potassium were added at a rate of 150 kg/fed, calcium super phosphate (15.5% P2 O5) and 50 kg/fed

potassium sulfate (48.5% K2 O), respectively. Three irrigations were added during growth by flooding system. In all experiments, weeds were controlled by hand as needed. Other cultural practices were kept constant for all the treatments.
Data Collection
Data on different agronomic traits were collected on both of plant and plot basis. Measurements and observation of examined characters were done on ten plants randomly chosen in the middle-row of each plot. Eight different traits were measured including grain yield and morphological traits and yield components, plant height, days to 50% heading, number of spikes per square meter, number of grains per spike, grain weight per spike, thousand grain weight and spike length. At maturity, grain yield of the three middle rows of each plot was determined and converted into tons per feddan.
BRIEF DESCRIPTION OF THE DIAGRAM
Fig.1: Intelligent Multiple Regression Analysis in Agriculture: Crop Productivity,
Rainfall, Fertility, Humidity, Temperature Using Machine Learning.
Fig.2: Intelligent Multiple Regression Analysis in Agriculture Prediction Location.
Fig.3: Crop Productivity, Rainfall, Fertility, Humidity, Temperature Using Machine
Learning.
Fig.4: Intelligent: Crop Productivity, Rainfall, Fertility, Humidity, Temperature
Using Machine Learning.
DESCRIPTION OF THE INVENTION General discussion:
Such research is susceptible to threats to validity, and potential threats to validity can be external, construct validity, and reliability (Šmite et al., 2010). The external validity and construct validity are addressed for this SLR study since the initial search string was broad, and the query returned a substantial number of studies: 567 publications in total.
The search string covered the whole scope of the SLR. For reliability of the SLR, the validity can be considered well-addressed since the process of the SLR has been described clearly and is replicable. If this SLR is replicated, it could return slightly different selected publications, but the differences would be a result of different personal judgments. However, it is highly unlikely that the overall findings would change.
Search-related discussion:
There is a possibility that valuable publications might have been missed. More synonyms could have been used, and a broader search could have returned new

studies. However, the search string resulted in a high number of publications indicating a broad enough search.
Analysis-related discussion:
Another issue that could be a threat to validity the way the analysis is conducted. For example, not all publications stated what kind of evaluation parameters were used, and sometimes just a few examples of features were explained. Thus, sometimes this information that is required to address the research questions could not be found in the paper.
This way, the data that was used to answer the research questions were derived from a few numbers of publications than a total of 50 selected publications. To get more information about the publications, the authors could potentially have been contacted, but this line of action was not feasible within the context of this research, and that might also not solve all the issues.
RQ1-Related (algorithms) discussion:
Linear Regression is the second most used algorithms, according to Table 5. Linear Regression is used as a benchmarking algorithm in most cases to check whether the proposed algorithm is better than Linear Regression or not. Therefore, although it is shown in many articles, it does not mean that it is the best performing algorithm. Table 5 should be interpreted carefully because “most used” does not mean the best performing ones. In fact, Deep Learning (DL), which is a sub-branch of Machine Learning, has been used for the crop yield prediction problem recently and is believed to be very promising. In this study, we also identified several deep learning-based studies.
There are several additional promising aspects of DL methods, such as automatic feature extraction and superior performance. We expect that more research will be conducted on the use of DL approaches in crop yield prediction in the near future due to the superior performance of DL algorithms in other problem domains.
Conclusion:
This study showed that the selected publications use a variety of features, depending on the scope of the research and the availability of data. Every paper investigates yield prediction with machine learning but differs from the features. The studies also differ in scale, geological position, and crop. The choice of features is dependent on the availability of the dataset and the aim of the research.

Studies also stated that models with more features did not always provide the best performance for the yield prediction.
To find the best performing model, models with more and fewer features should be tested. Many algorithms have been used in different studies. The results show that no specific conclusion can be drawn as to what the best model is, but they clearly show that some machine learning models are used more than the others. The most used models are the random forest, neural networks, linear regression, and gradient boosting tree. Most of the studies used a variety of machine learning models to test which model had the best prediction.
The present invention is a method, program storage device readable by a machine and crop yield predicting device for predicting crop yield. Historical satellite and meteorological data may be used to derive empirical equations to forecast crop yield. Specifically, NDVI, SM, ST and Rainfall data may be fitted to predefined empirical equations using piecewise linear regression and optimization of least square loss function through a non-linear Quasi-Newton method. A non-linear and iterative method can help minimize loss function to yield values for coefficients of the empirical equations.
The derived empirical equations can be used to forecast crop yield of any crop of choice based on data obtained before harvest. The piecewise linear regression and Quasi-Newton method can also be applied to other states, provinces, territories, commonwealths, countries, etc., where crop production may be dependent on weather and climatic conditions.
The invented machine learning algorithms have been applied to support crop yield prediction and we performed a Systematic Literature Review (SLR) to extract and synthesize the algorithms and features that have been used in crop yield prediction studies.
Based on our search criteria, we retrieved 567 relevant studies from six electronic databases, of which we have selected 50 studies for further analysis using inclusion and exclusion criteria. We investigated these selected studies carefully, analyzed the methods and features used, and provided suggestions for further research.
The invention is a temperature, rainfall, and soil type, and the most applied algorithm is Artificial Neural Networks in these models. After this observation based on the analysis of machine learning-based 50 papers, we performed an additional search in electronic databases to identify deep learning-based studies,