Comp. Intelligence & Neural Nets for Activity Classif.

Assignment: Computational Intelligence and Neural Networks for Activity Classification

Learning Outcomes

On the completion of this assignment you will be able to:

Critically appraise a comprehensive/detailed understanding of the computational intelligence domain.
Design and develop computational intelligence software artefacts.
Critique and contextualise emerging research in the area of computational intelligence.

Edge

The Cardiff Met EDGE supports students in graduating with the knowledge, skills, and attributes that allow them to contribute positively and effectively to the communities in which they live and work.

This module assessment provides opportunities for students to demonstrate development of the following EDGE Competencies:

ETHICAL	Students will be required to consider Ethical implication of their analysis and follow the necessary ethical approval processes while addressing problems associated with the Product Activity Classification assessment.
DIGITAL	Students will be required to demonstrate digital skills in the collation of data and analysis for their project. Hence the use of MATLAB mobile to collect sensor data, and python to carry out some more pre-processing and analysis.
GLOBAL	Students will demonstrate an awareness of the global context and apply this to their Product Activity Classification assessment
ENTREPRENEURIAL	Students will also demonstrate their developed entrepreneurial through working under their own initiative, formulating and presenting recommendations in order to solve an authentic and complex problem associated with the module.

Assessment Requirements / Tasks (include all guidance notes)

Today, a lot of Computational Intelligence artefacts power various industries and sectors globally. The world is awake to an exponential growth in technological advancements in intelligence systems that leverage various forms of data.

The aim of this assignment is to provide students with an opportunity to gain experience using specialist languages, software and development packages, to investigate the application of computational intelligence (Neural Networks) in the ubiquitous system's domain.

IBM projected that the Internet of Things (IoT) market will grow from an installed base of 15 billion devices in 2015 to about 75 billion by 2025.

The overall task of this assignment is for you to build THREE neural networks for activity classification which should be initially trained on your training dataset, and then evaluated on your test dataset.

You are to use this Product activity data which captures a number of activities. Click this Link to access the Chatty-Dataset .

You are required to Pre-process the data and carry out some exploratory data analysis before building your models. Pay attention to these sensors: "Acceleration", "Magnetic Field", "Orientation" and "Angular Velocity" which were used to capture data. You are to discuss them in your report. Furthermore, as mentioned earlier you are expected to build THREE Artificial Neural Networks to classify the activities in the Chatty-Dataset. In your report, discuss the best performing model and highlight their hyperparameters and accuracy.

Your report should contain:

An introduction which describes the dataset used
Discuss the sensors ("Acceleration", "Magnetic Field", "Orientation" and "Angular Velocity)
Discuss your model creation,
hyperparameters and model performance.
You are also expected to use appropriate plots/ graphs and provide a short description. Exploratory Data analysis using Visuals (plots/ graphs)
Finally, you need to provide a concluding section reflecting the knowledge gained from your analysis.

Get instant help from 5000+ experts for