KNN Application
Name
Institution Affiliation
Date
Executive summary
Data mining methods have been commonly used to extract rich information from clinical databases. The data mining findings indicate that deep supervised classification methods, which can be used to build models representing essential data groups where even the transferase is involved in this project of the classification model. KNN is a very straightforward technique, most popular, efficient, and effective information-processing algorithm. KNN is a transparent forward classification algorithm, where observations are categorized according to the category of their nearest neighbor. Browsing history is of significant intensity in nature.
If the data set contains redundant information features, the group can yield less reliable data. Data mining is the method of extracting relevant, credible information from large quantities of data. This has become extremely valuable, as everyday data has grown tremendously. Software mining is an integrated component of KDD, which contains a series of transition phases from before the processing of data to processing and analysis of data mining algorithms. KNN is used in the classification of the diabetes dataset to check if the patient has diabetes or not. It is found that the model produced has an accuracy of 73%.
Introduction
The primary function of the data mining KNN algorithm is classification, connection, and grouping. Categorization is a widespread concern that covers a wide variety of applications. Data analysis methods have been applied to several medical fields to enhance diagnosis and treatment. Many healthcare providers are faced with a significant problem: delivering high-quality services such as accurately medical diagnosis and offering treatment at a fair cost. Data mining methods address a variety of standard and relevant questions related to the patient. KNN is among the most common classification techniques developed by Hodges and Fix (Chen, 2018). With no extra data, the guidelines for identification are created by the training images together. Adaptive methods are used to solve an issue that cannot increase the accessibility of the proposed algorithm (Guo et al., 2013). Genetic algorithms (GA) are coding contractor management on the evolutionary line. GA Has played a critical role in a variety of computer technologies.
Diabetes and heart disease remains the primary cause of death in the world, which is one of the main contributors to the burden of the disease in emerging countries such as Malaysia. A few studies in Madhya Pradesh have shown cardiac and diabetes disease is the leading cause of death, taking account for 32% of fatalities, as rising as Canada and the United States. There is, therefore, a need to design and create clinical decision making for the categorization of cardiac disease (Liu & Liu, 2015). In this paper, a prediction method uses KNN and evolutionary algorithms to predict a person’s diabetes status.
Model/methods
The dataset in this analysis involves finding out whether a patient has diabetes or not. The method used in this problem is the K nearest neighbor.
K nearest neighbor procedure.
The algorithm can be summarized as:
- A value k is specified, and a new sample selected.
- From the database, an entry k is selected closer to the new sample.
- The most common classification of the entries is selected.
- Then the new sample is assigned to the classification.
The diabetes dataset from the Kaggle website is used for analysis. The variables in the dataset include:
Glucose, Pregnancies, Bloodpressure, Skinthickness, Insulin, BMI, Age, and outcome.
The objective of this analysis is to train a model using a historical dataset where the model will be applied in a real-world scenario.
Solution and Analysis
In the analysis, the dataset will be cleaned for further analysis. The dataset will first be split into training and testing datasets. The training dataset will be used to train the model, while the testing dataset will be used to test the accuracy of the model developed.
After splitting the dataset into testing and training set, the dataset is then fit into the model.
Using the different values of k, the accuracy of the model is tested.
Based on the results above, the value of k of seven gives better accuracy of the model. Now using the value of k of seven, a model is fit.
This gives an accuracy of 73.05%. This means that out of the total patients, the model will predict correctly approximately 73.05% of the cases.
In conclusion, machine learning is vital in solving the current world problems. However, it is advisable to use a considerable amount of datasets to ensure the model is well trained hence the accuracy of the model. Algorithms such as KNN help classify data into a positive or negative category. Where, in this case, it helps identify whether a patient has diabetes or not.
References
Chen, J. H. (2018). KNN based knowledge-sharing model for severe change order disputes in construction. Automation in Construction, 17(6), 773-779.
Guo, G., Wang, H., Bell, D., Bi, Y., & Greer, K. (2013). KNN model-based approach in classification. In OTM Confederated International Conferences” On the Move to Meaningful Internet Systems” (pp. 986-996). Springer, Berlin, Heidelberg.
Liu, Q., & Liu, C. (2015). A novel locally linear KNN model for visual recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1329-1337).