AU2020100954A4 - Decision making system for crop-livestock farms using machine learning algorithms - Google Patents

Abstract

DECISION MAKING SYSTEM FOR CROP-LIVESTOCK FARMS USING MACHINE LEARNING ALGORITHMS Abstract: The machine learning have augmenting with decision making innovations and performance calculation to make modern opportunities used for information seriously science within the different disciplinary Agri-technologies. We display a comprehensive show for crop-livestock farms to commit the claim of machine learning in agricultural fabrication systems. Analysis is conducted and demonstration were sent in (a) crop management, counting purpose on yield forecast, disease discovery, weed detection crop quality, and species acknowledgment; (b) livestock management, counting applications on creature welfare and livestock production; (c) soil management; and (d) water management. The examined and classification of the decision-making framework for crop livestock is illustrate how farming will advantage from the machine learning advances. Machine learning algorithm to applying for sensor information, cultivate management frameworks are developing into genuine time and forged impending authorized programs that give wealthy counsel and approaching for farmer decision support and action. 11Page DECISION MAKING SYSTEM FOR CROP-LIVESTOCK FARMS USING MACHINE LEARNING ALGORITHMS Machine Learning in Agriculture Supply Chain Group|| GroupIII Production Phase Processing Phase Weather Disease Demand Production Prediction Prediction Mgmt Planning Weed Livestock Quality Prediction Mgmt Mgmt Nutrient Crop Mgmt Harvest Fig2: Machine Learning Monitoring and Controlling

Description

DECISION MAKING SYSTEM FOR CROP-LIVESTOCK FARMS USING MACHINE LEARNING ALGORITHMS

Machine Learning in Agriculture Supply Chain

Group|| GroupIII

Production Phase Processing Phase

Weather Disease Demand Production Prediction Prediction Mgmt Planning

Weed Livestock Quality Prediction Mgmt Mgmt

Nutrient Crop Mgmt Harvest

Fig2: Machine Learning Monitoring and Controlling

DECISION MAKING SYSTEM FOR CROP-LIVESTOCK FARMS USING MACHINE LEARNING ALGORITHMS

Field of Invention Field of Invention Related to crop-livestock farms using Machine Learning

Soil management For masters included in agriculture, soil may be a heterogeneous characteristic asset, with complex forms and dubious components. Its temperature alone can grant experiences into the climate alter impacts on the territorial yield. Machine learning algorithms think about evaporation forms, soil dampness and temperature to get it the flow of biological systems and the impingement in farming.

Water Management Water management in agribusiness impacts hydrological, climatological, and agronomical adjust. So distant, the foremost created ML-based applications are associated with estimation of every day, week after week, or month to month evapotranspiration permitting for a more successful utilize of water system frameworks and forecast of day by day dew point temperature, which makes a difference distinguish anticipated climate wonders and gauge evapotranspiration and dissipation.

Background and prior art of the invention: Crop management Yield Prediction Yield expectation is individual of the foremost important and prevalent topics in exactness agribusiness because it characterizes yield diagram and inference, coordinating of edit provide with request, and crop management. State-of the-art approaches have gone distant past straightforward expectation based on the historical information, but consolidate computer vision advances to supply information on the go and comprehensive multidimensional examination of crops, climate, and financial conditions to create the foremost of the surrender for agriculturists and populace.

11Page

Crop Quality The precise discovery and categorization of crop quality definitions can increment product cost and decrease squander. In comparison with the human specialists, machines can make utilize of apparently insignificant information and interconnects to uncover modem qualities playing part within the by and large quality of the crops and to distinguish them.

Disease Detection Both in open-air and greenhouse situation, the foremost broadly utilized hone in bug and disease control is to consistently splash insect repellent over the trimming region. To be successful, this approach requires noteworthy sums of pesticides which comes about in a tall money related and noteworthy natural taken a toll. ML is utilized as a portion of the common accuracy horticulture administration, wherever agro-chemicals input is focused on in terms of instance and influenced plants.

Weed Detection Separated from infections, weeds are the foremost imperative dangers to crop generation. The greatest issue in weeds battling is that they are troublesome to identify and separate from crops. Computer vision and ML calculations can move forward location and separation of weeds at low cost with no natural subject and area impacts. In future, these innovations will drive robots that will annihilate weeds, minimizing the require for herbicides.

Livestock management

Livestock Production Comparable to crop administration, machine learning gives exact forecast and inference of cultivating limitation to optimize the financial proficiency of livestock fabrication systems, such as cattle and eggs production. For illustration, weight foreseeing frameworks can assess the longer term weights 150 days earlier to the butcher day, permitting agriculturists to alter diets and conditions individually.

21 P a g e

Animal Welfare

In present-day setting, the livestock is progressively treated not fair as food holders, but as creatures who can be despondent and depleted of their life at a cultivate. Creatures behavior classifiers can interface their chewing signals to the require in slim down changes and by their development designs, counting standing, moving, bolstering, and drinking, they can tell the sum of push the creature is uncovered to and anticipate its defenselessness to infections, weight pick up and production.

Objective of the invention:

The main objective machine learning algorithms for agriculture to make the model visualize various situations and take the action accordingly. ML algorithms can be urbanized to assist and resolve which hybrids have the likelihood of achieving maximum yield impending in every environment.

Statement of the invention:

Machine Learning Calculation is proposed to screen and control the crop -livestock cultivate forecast condition of the prepared information set. The decision making mechanism are utilized to classify the data such as labeled and unlabeled data. The decision making are communicated by itself since machine learning is utilized to put through the crop-livestock farms. The machine learning subset of manufactured insights is to run the show the anticipated information around farms and share the data through agriculturists. The agriculturists can be utilized as interface between the machine learning and decision making mechanism. It is associated to the Al based different applications can too give the subtle elements to develop the edit in superior way and improve the yield. This strategy could be a modern move toward for fabrication of agricultural crop management. Exact and opportune figures of crop production are important for imperative approach decisions like import-export, estimating promoting dispersion etc. which are subjected by the directorate of economics and statistics. The machine learning gives low cost and adaptable connector for observing and controls the crop - livestock farms management and machine learning algorithms would be competent to recommend the leading crop to develop. The part of machine

3|Page learning algorithms in giving real-time systematic bits of knowledge for positive data-driven decision-making within the Agriculture.

BRIEF DESCRIPTION OF THE SYSTEM OF DRAWINGS Fig 1: Data flow Diagram Fig 2: Machine Learning Monitoring and Controlling Fig 3: Machine learning with support of Decision Making

DETAILED DESCRIPTION OF THE SYSTEM The machine learning expressions and characterization normally, Machine Learning strategies includes a learning prepare with the intention to memorize from "experience" (preparing information) in the direction of execute an assignment. Information in machine learning comprises of a rest of illustrations. More often than not, a person case is depicted by a rest of qualities, moreover acknowledged as highlights or factors. The highlighting can be (numbers, genuine number, etc.), double (i.e., or 1), ordinal (e.g., A+ or B-), or ostensible (identification) numeric. The execution of the machine learning demonstrate during a particular assignment is deliberate by an execution of metrics that moved forward with involvement in excess of time. In the direction of calculating the execution of machine learning models and computations, different quantifiable and numerical models are exploited. The subsequent of the conclusion to the learning prepare, the organized demonstrate can be employed to classify, foresee, or cluster unused cases (testing data) utilizing the encounter gotten amid the preparing prepare. Figure 1 appears an commonplace machine learning approach. Sensors 2018, 18, x FOR PEER Audit 4 of 31 metrics that progressed with experience over time. Toward calculate the execution of machine learning models and computations, different quantifiable and scientific models are employed. Later than the conclusion of learning handle, the equipped show can be utilized to classify, foresee, otherwise cluster unused cases (testing data) utilizing the involvement gotten amid the preparing prepare. Figure 1 appears a normal machine learning advance. Machine learning assignments be ordinarily confidential into distinctive wide type depending on the learning models (classification, relapse, clustering, and dimensionality diminishment), learning sort (supervised/unsupervised), otherwise the learning models utilized to execute the chosen task.

41Page

Claims (1)

1. DECISION MAKING SYSTEM FOR CROP-LIVESTOCK FARMS USING MACHINE LEARNING ALGORITHMS

   We claim that,

   • A monitor and controlling system consist of machines learning along with

   number of decisions making such as labeled and unlabeled data etc.

   • Machine Learning Algorithm is proposed to monitor and control the crop

   livestock farm prediction condition of the trained data set.

   • This technique is an innovative approach for fabrication of agricultural crop

   management. Accurate and appropriate forecasts of crop production are

   required for important procedure decisions like import-export, pricing

   marketing distribution etc.

   • A machine learning algorithm with touch screen such as smart phone, laptop, tablet etc. with high speed internet or Wi-Fi for monitoring and controlling

   purpose.

   • The part of Machine Learning algorithms in given that real-time analytic

   impending for practical data-driven decision-making in the Agriculture.

   • Farmers can be used as interface between the machine learning and decision

   making mechanism.

   1|Page

   DECISION MAKING SYSTEM FOR CROP-LIVESTOCK FARMS USING MACHINE Jun 2020

   LEARNING ALGORITHMS

   New Examples 2020100954

   Training Data Machine Learning Classification/ (Labeled/unlabeled) Algorithm Prediction rule

   Predicted Outputs Figure 1. Data Flow Diagram

   DECISION MAKING SYSTEM FOR CROP-LIVESTOCK FARMS USING MACHINE Jun 2020

   LEARNING ALGORITHMS

   Machine Learning in Agriculture Supply Chain 2020100954

   Group II Group III

   Production Phase Processing Phase

   Weather Disease Demand Production Prediction Prediction Mgmt Planning

   Weed Livestock Quality Prediction Mgmt Mgmt

   Nutrient Crop Mgmt Harvest

   Fig2: Machine Learning Monitoring and Controlling

   Outcomes Crop and livestock products, income, environment impacts

   Fig3: Machine Learning with support of Decision Making