ER4165 Advanced Topics in Machine Intelligence

ER4165 Advanced Topics in Machine Intelligence

Quality control (QC) is a process through which a business seeks to ensure that product quality is maintained or improved. Quality control involves testing units and determining if they are within the specifications for the final product. In this design study, you will provide a detailed machine learning solution that will be utilized to evaluate the quality of the product based on a previously collected dataset. The solution includes various tasks related to the design, including features selection, dimensionality reduction, classification, and the quantitative evaluation of the provided solutions. This task will test your ability to: • Design ML methods and processes necessary to deploy an artificial intelligence system for quantitative evaluations of a mechatronic system. • Recognize software design challenges behind implementations of machine learning algorithms. • Design and optimise software to meet specified requirements. • Design and provide a working solution for Mechatronic System Quality. These correspond to point 1, 2, 3, and 4 of the module learning outcomes. Design Study Descriptions and Objectives: The objective of this task is to identify which system is performing within the required specifications. The produced system can be classified as a Pass or Faulty system based on six measurements related to the system vibration, speed response, stability, transmission response, power consumption, and shear stress. These features have been collected for a large population of around 4K readings that have been logged. The TrainingSet data is available as TrainingSet.mat (Matlab format file) on the Blackboard. This set consists of readings of the measurements with their corresponding ground truth (class). The first six columns are the system measurements, while the seventh column is the class. The class for each reading is either Pass indicated by ‘logic 0’ or Faulty indicated by ‘logic 1’. The training dataset contains 3800 readings. The additional TestData.mat file is also available from the Blackboard. That file contains 200 additional readings, with recorded values of the six features. However, althoughthe fault ground truth labels are known, you will not have access to that information before the submission deadline. Design Study Requirements You are asked to design a system for recognition between a faulty system and a pass system. In your design, you should: consider a need for feature selection or feature dimensionality reduction; if necessary, apply a suitable feature selection/reduction algorithm; select a machine learning (ML) algorithm suitable for the task; training of the selected ML algorithm; quantitatively evaluate the design system.

ER4165 Advanced Topics in Machine Intelligence

Quality control (QC) is a process through which a business seeks to ensure that product quality is maintained or improved. Quality control involves testing units and determining if they are within the specifications for the final product. In this design study, you will provide a detailed machine learning solution that will be utilized to evaluate the quality of the product based on a previously collected dataset. The solution includes various tasks related to the design, including features selection, dimensionality reduction, classification, and the quantitative evaluation of the provided solutions. This task will test your ability to: • Design ML methods and processes necessary to deploy an artificial intelligence system for quantitative evaluations of a mechatronic system. • Recognize software design challenges behind implementations of machine learning algorithms. • Design and optimise software to meet specified requirements. • Design and provide a working solution for Mechatronic System Quality. These correspond to point 1, 2, 3, and 4 of the module learning outcomes. Design Study Descriptions and Objectives: The objective of this task is to identify which system is performing within the required specifications. The produced system can be classified as a Pass or Faulty system based on six measurements related to the system vibration, speed response, stability, transmission response, power consumption, and shear stress. These features have been collected for a large population of around 4K readings that have been logged. The TrainingSet data is available as TrainingSet.mat (Matlab format file) on the Blackboard. This set consists of readings of the measurements with their corresponding ground truth (class). The first six columns are the system measurements, while the seventh column is the class. The class for each reading is either Pass indicated by ‘logic 0’ or Faulty indicated by ‘logic 1’. The training dataset contains 3800 readings. The additional TestData.mat file is also available from the Blackboard. That file contains 200 additional readings, with recorded values of the six features. However, althoughthe fault ground truth labels are known, you will not have access to that information before the submission deadline. Design Study Requirements You are asked to design a system for recognition between a faulty system and a pass system. In your design, you should: consider a need for feature selection or feature dimensionality reduction; if necessary, apply a suitable feature selection/reduction algorithm; select a machine learning (ML) algorithm suitable for the task; training of the selected ML algorithm; quantitatively evaluate the design system.

Manner Wilfrid xanaxmd.com dreamt smartly valimeds.com endure beaming ativanmd.com security cell