Application of Artificial Intelligence In The Detection Of Plant Diseases (Clubroot)

Indonesia is a tropical country with diverse flora and fauna stretching from west to east. One of the cabbage plants grown in Indonesia has many benefits for human health. Cabbage plants are susceptible to diseases such as pests and pathogens. One of them is clubroot disease, which causes the plant not to grow and develop. Based on the above problems, it is necessary to have a system to support farmers in the scientific field of expert systems that uses the Certainty Factor method to diagnose the nature of clubroot disease. It is hoped that this system can provide information more quickly on whether or not cabbage plants have clubroot disease. The results of this study represent an integrated system capable of solving problems in cabbage plants, especially in the diagnosis of clubroot disease. It is hoped that the applied system can be further developed together with technological developments.


Introduction
Cabbage is a vegetable plant belonging to the Brasccae Oleracae var. capitat L family, often called cabbage or cabbage. Cabbage vegetables contain valuable ingredients such as vitamin A, vitamin B, vitamin C and vitamin E, as well as many minerals, namely potassium, calcium and phosphorus, and contain anticancer compounds [1]. However, efforts to grow cabbage plants are often attacked by various pests and pathogens that cause plant and crop diseases. One of these is clubroot disease, which occurs in cabbage plants [2]. Clubroot disease or clubroot is a disease caused by the soil-borne pathogen Plasmodiophora brassicae Wor. This pathogen causes swelling of the root tissue, which can impair root function such as the transfer of nutrients and water from the soil to the leaves [3].
To overcome this, one needs an expert or a cabbage plant disease expert. Meanwhile, not all cabbage disease experts can help in overcoming these problems every time. a system that can help an expert or professional to find out the disease that has occurred more easily. One of these is the use of expert system knowledge. A universal expert system is a scientific field that can be used in diagnosing a disease and is implemented on a computer. According to the understanding expressed by Ignizo, the expert system is a procedure related to a specific container whose level of knowledge can be compared with that of an expert [4]. In another study, an expert system was used to diagnose onion plant diseases, with the result that the expert system was able to search for symptoms, diseases and solutions by looking for answers to questions posed by the system [5].
In the expert system, there is a method, namely Certainty Factor (CF), which is used in this study. The Certainty Factor method is used to deal with uncertainty in the creation of MYCIN. The certainty factor (CF) is a clinical parameter value given by MYCIN to indicate the level of confidence. The certainty factor method is used to track the input of damage symptoms and then determine the density value (confidence) for each damage according to the symptoms. After the identity value is determined, the combination of rules can be determined. In MYCIN, there are rules for combining symptoms in a certainty factor rule defined as equation [6].

Research Methodology
In the concept of writing, the system design method is one of the most important elements in research.
In the system design method, especially in software, various methods can be applied, including the waterfall algorithm or the waterfall algorithm used in this study [7].
In the structure of the expert system, there are two main components, namely the development environment and the consulting environment. The development environment is used for the process of developing an expert system both in terms of producing the components and the knowledge base. consult an expert (expert) [8] [12]. Expert systems have the ability to recommend a set of actions or user behaviour to run a correct and accurate correction system. In doing so, this system also uses reasoning skills to draw conclusions based on available data and facts. [11] An expert system aims to produce a conclusion or description from a set of raw data (raw data). Decision-making is based on observations, starting with image analysis, word recognition through speech, signal interpretation and the ability to project the consequences of certain situations and conditions [13].

System Algorithm
In solving problems encountered in diagnosing clubroot disease in cabbage, knowledge representation is the method used for encoding expert system knowledge. The following is a system algorithm for completing an expert system for diagnosing clubroot disease in cabbage plants, including [14]: 1. The determination of disease and symptom data. Flowchart is a flowchart that contains an algorithm for diagnosing a disease with symptoms or steps present using the Certainty Factor method and the detailed process sequence. The design of a flowchart consists of three parts, namely input, process and output [1]. The basic purpose of flowcharts is to describe a phase of problem solving in a simple and sequential manner using predetermined symbols or standards. Here is a picture of the flowchart [15]: The first step in the method of certainty factors is: 1. Assign a value to each symptom in the knowledge base 2. Calculate the certainty factor value of each user response given 3. Calculate the certainty factor value of the results of each user response multiplied by the symptom value using the certainty factor combination.
The results of the user responses from the expert system process are then processed to calculate the certainty value using the certainty factor method. The certainty factor is a method for proving the untrustworthiness of an expert. Certainty is determined by giving a weight to each symptom selected by the user. The certainty factor method thus provides a confidence value [13].

Symptoms and Disease Data
The success of an expert system results from the knowledge of experts with assumed experts, and how to manage the knowledge gained from the results of the survey and then create it as a disease table to help diagnose clubroot disease in cabbage plants can be seen in the The underside of the plant turns yellow The process of collecting from experts, it was found that for the high category of cabbage plant diseases had symptom characteristics given the initials X005, X006, X008, X009 and X010. For the medium category, the symptoms are X003, X004, X006, X007 and X009, while the low category is X001, X002, X003 and X004.

Measure of Belief (MB) and Measure of Disbelief (MD) Weight Value
Each symptom of the disease has a weight value or value that contains certainty and uncertainty as follows: Based on the grouping of symptoms and the determination of the cabbage disease category, the value of Measure of Belief and Measure of Disbelief was obtained from each category (High, Medium, and Low). This data was obtained from various journals and previous research. The value of cabbage disease in the high category has a Measure of Belief value of 0.7, 0.7, 0.7, 0.8, 0.9 in the high category, 0.6, 0.7, 0.7, 0.7, 0.9 in the medium category, and 0.7, 0.5, 0.6, 0.7 in the low category.

Rules
Rules are usually written in the if-then form (If....Then....). The following rules can be taken from the above disease

Manual Calculation of Certainty Factor Method
The following is an example of a case study of the manual calculation of the certainty factor method:  The next step is to combine the values from CF to measure the degree of certainty in diagnosing the symptoms of Clubroot disease. Below is the formula for combining the CF values: CF Combine (CF,CF2) = CF1 +CF2*(1-CF1) 1.
Plasmodiophora = 0,982 * 100% = 0,98,2 % The results of the calculations and diagnoses are explained on the basis of experts who have an assessment of the certainty factor in this case, the cabbage plant has clubroot Plasmodiophora brassicae wor disease (High) with a CF value of 99.4%, which means that the degree of confidence in the disease is 99% affected. clubroot disease.

Application and Implementation
Implementation is the phase in which the created system is executed or operated. This phase explains how you can operate the created system.

User Main Page
The following is a display of the main user page, namely: The user's main menu display is the start display or the user's first display in the cabbage disease diagnosis expert system. On the user's main menu display, there is also a menu with options for the disease diagnosis pages, expert profiles and admin logins.

Diagnosis Page
The following is a display of the user diagnosis page, namely: Figure 4. User Diagnosis Page. The menu display on the diagnosis page where the user diagnoses cabbage plant diseases by selecting which symptoms occur in cabbage plants. The user confidence menu display shows the questions for each symptom that occurs in cabbage plants. In this view, the user has to fill in each question and select one of the beliefs. There are 10 choices, namely: probably, very probably, almost certainly and definitely.

Diagnostic Results Page
The following is a display of the user diagnostic results page, namely:

Conclusion
The following conclusions can be drawn from the results of the investigation of clubroot disease using the Certainty Factor method: An expert system for plant diseases using the Certainty method has been successfully built and displays the symptoms, the results of calculations also using the Certainty factor, the results of the diagnosis and the solution of the disease. The system created is able to provide information about diseases in cabbage plants in a simple way, using the symptoms selected in the system according to the conditions to which the cabbage plants are exposed.