Nasaie Zainuddin 1 , Najua Tulos 1 , Nurain Yusof 1 , Nor Juliana Mohd Yusof 1 , Mohd Norazmi Nordin 2 1823 Turkish Online Journal of Qualitative Inquiry (TOJQI) Volume 12, Issue4, July , 1823- 1832 Research Article Machine Learning-Based Simulation In Remote Sensing Contexts Nasaie Zainuddin 1 , Najua Tulos 1 , Nurain Yusof 1 , Nor Juliana Mohd Yusof 1 , Mohd Norazmi Nordin 2 1 faculty Of Applied Sciences, Universiti Teknologi Mara, Shah Alam, Selangor, Malaysia 2 cluster Of Education And Social Sciences, Open University Malaysia Abstract The Current Study Involved Two Proposed Algorithms: K_Wic And K_Cio. The Algorithm K_Wic Generates The Good Initialization Center Set And Supports For Speeding Up Execution Of Remote Sensing Image Clustering. The Algorithm K_Cio Is Used To Cluster Remote Sensing Images With Adding Context Information Of Each Pixel. The Test Results Show That The Algorithms K_Wic, K_Cio And K_Wic_Cio (A Combination Of K_Wic And K_Cio) Can Be Used Effectively For Remote Sensing Image Clustering. Besides, Due To The Nature Of The Wavelet Transform, The Value Domain Of The Output Data Is Changed. Specifically, Image Doesn’t Belonging To The Domain [0,255]. Therefore, We Proposed An Improvement Of Wavelet Transformation To Still Ensure That The Domain Of The Output Data Belongs To The Domain [0,255], Suitable For Image Data. In Future Work, We Will Continue To Study The New Context Information And The New Algorithms. 1 Introduction Clustering Is A Process Used To Extract The Main Features Of Background Objects By Defining Corresponding Regions [1]. The Task Of Image Segmentation Is From The Initial Multi-Spectral Image, Proceeding To Gather Pixels With The Same Properties (Color, Shape, Texture) Into The Same Cluster To Divide Into Regions And Clusters. Currently, There Are Many Different Partitioning Methods Such As: Morphological Methods, Kmeans Methods, Limited Gaussian Mixing Model (Fgmm), Separation And Integration, Markov Models. Most The Methods Only Use Pixel Intensity Characteristics To Perform Clustering. At Present, Some Algorithms Use More Contextual Information In The Process To Reduce The Complexity Of Segments [2]. In [3], The Authors Also Combined Fuzzy Clustering Algorithms And Other Gray Level Adjustment Expressions To Enhance Medical Imaging Contrast. In [4], The Authors Used The Local Approach To Enhance The Contrast Of Remote Sensing Images. In [5], The Authors Improved The Kmeans Algorithm To Use The Cluster Center Instead. Every Aspect Requires Efficient Management (Abdul Jalil Et Al., 2021; Mohd Noh Et Al., 2021; Mustafa Et Al., 2021; Roszi Et Al., 2021; Tumisah Et Al., 2021). If It Is Managed Well, Various Problems Can Be Avoided (Irma Et Al., 2021; Suzana Et Al., 2021; Rohanida Et Al., 2021; Nazrah Et Al., 2021; Shahrulliza Et Al., 2021). Fuzzy Cmeans Algorithm Is Highly Appreciated For Image Processing With The Application Of Fuzzy Clustering. It Is Very Important That Fuzzy Cmeans Allows Control Over The Number Of Clusters. However, The Execution Speed Of This Algorithm Is Very Slow. With The Large Images Like Remote Sensing Images, The Speed Is Even Slower. In Addition, The Membership Matrix Is A Major Obstacle For This Algorithm To Perform With Large Images Like Remote Sensing Images. In Addition, According To [6], Kmeans Loses The Contextual Characteristics (Neighboring Information) Of Each Pixel When Only Intensity Characteristics Are Considered. In [7], The Authors