Artificial Neural Networks

Learning Outcome 1: Acquire a deep knowledge of the constitutional concepts of artificial neural networks including theirbiological inspiration.

Learning Outcome 2: Apply and compare the different architectures and learning approaches available in neural networksystems.

Learning Outcome 3: Design and develop different neural network models applying appropriate learning approaches forreal world applications.

Learning Outcome 4: Use the available neural network simulators, develop solutions to real-world problems and appraisetheir limitations.

Learning Outcome 5: Critically evaluate the trends in neural network developments.

Task:
In this assignment, you will select one task such as classification (including object recognition), regression, clustering, forecasting, and recommendationfor a problem inspired from the real world and explore how to best apply (deep) neural networks to solve it.

The main purpose of this assignment is to:
• Test the understanding of fundamental concepts of neural networks and their applications.
• Perform appropriate preparation of a dataset and evaluate the performance of different neural networkson the chosen dataset.
• Gain practical experience in using neural network learning algorithms for solving a real-life problem.
• Demonstrate the ability to critically evaluate the results of different learning algorithms.

Organisation of the report and marking scheme:
o Cover page: the title of the project

o Section 1. Introduction
Should include the description of the chosen real-life problem and its significance (1-2 pages)

o Section 2. Related work
Should explain and properly cite the (most notable) works already done related to the selected problem. The references need to be from journal papers not just from websites.

o Section 3. Dataset
Should include the description of the publicly available dataset and its link (1-3 pages). Please describe the processes you use to clean and formalize the data set for your use.

Section 4. Method(s)
Should describe the appropriate neural network(s), together with the learning algorithm(s) and other parameters, tried to solve the problem. Please provide the reasoning behind selecting the type of the neural network and parameters.

o Section 5. Experimental results
Should explain the rigor experiments conducted and compare theirdetailed results.

o Section 6. Discussion and future work
Should present the summary of the findings (1-2 pages). Also provide what future work can be conduced on this project.

o References
List of references cited in the other sections, in the APA style.

Attachment:- Artificial Neural Networks.rar