Full screen

Share

Show pages

Carlos Alberto Bustamante Gaytán
Neural network-based feedback and support interface
RESearch project conceptualization
22/11/2022
Want to make interactive content? It’s easy in Genially!

Over 30 million people build interactive content in Genially.

Check out what others have designed:

Transcript

Carlos Alberto Bustamante Gaytán

Neural network-based feedback and support interface

RESearch project conceptualization

22/11/2022

09.potential outcomes

10.references

07.schedule

08.swot analysis

06.resources

05.conceptual methodology

04.Hypothesis

03.research questions

02.aim and objectives

01.background

CONTENt

01

background

01. background

  • Machine Learning
  • Deep Learning
  • Supervised learning
  • Unsupervised learning
  • Neuron
  • Artificial neural network
  • Weights
  • Bias/Threshold
  • Cost function

key concepts

DEFINITION

Neural networks or artificial neural networks are a subset of Machine Learning and are the basis of Deep Learning algorithms. Their name and structure are inspired by the human brain and mimic the way biological neurons connect. They are composed of layers of nodes, which have within them an input layer, one or more hidden layers, and an output layer. If the output of any node is above the threshold value (bias), that node becomes active, sending data to the next layer of the network.

01. background

working principle

01. background

01. background

LITERATURE REVIEW

02

AIM AND OBJECTIVES

02. AIM AND OBJECTIVES

OBJECTIVES

  • Review the methodology of similar papers in high impact journals during the initial phase of the project.
  • Collect data generated by students during laboratory practices in the first two semesters (initial phase of the project).
  • Explore and prepare the data generated, so that they can be used for the neural network-based model (second phase of the project).
  • Develop the machine learning neural network-based model using evaluation metrics to select the best type of neural network.
  • Develop the user interface based on the Delta ASDA software interface taking into account functionality, accessibility and ease-of-use.
  • Evaluate the interface and model during the final phase of the project in real time laboratory practices.
  • Deploy the interface and the model until the expected laboratory performance is achieved.

AIM

Teaching new software tools in engineering has a learning curve, so performing exercises or lab practices may take longer than expected initially. The aim of the thesis is to develop a neural network-based interface with generated data by the students that allows finding the most common errors, providing feedback and reducing the time to perform the lab practices.

03

RESEARCH QUESTIONS

What type of Machine Learning algorithm best suits the generated data? Supervised learning, unsupervised learning or reinforcement learning?

How to generate real-time labeled data for future project refinements?

How can the user interface be designed to avoid being just a sequential (do as I say) lab practice?

What are the implications of using an interface based on neural networks (Education 4.0) as opposed to traditional self-guided education during laboratory practices?

What type of neural network can best fit the type of data generated during this and the next semester?

03. RESEARCH QUESTIONS

04

HYPOTHESIS

A neural network-based interface will reduce the software learning curve significantly while providing support and feedback in the realization of laboratory practices.

04. HYPOTHESIS

05

CONCEPTUAL METHODOLOGY

What type of neural network can best fit the type of data generated during this and the next semester?

05. METHODOLOGY

06

Resources

Time

Financial

Material

Human

  • Short period of time, less than two years to develop the project
  • Funding available for the acquisition of better equipment
  • Computers (laboratories)
  • Software (free)
  • Electric equipment (laboratories)
  • Cloud server (TBD)
  • Me
  • Thesis advisers
  • Grant holders

06. resources

07

schedule

5 stages

  1. Project definition/initiation
  2. Project research
  3. Project development
  4. Project evaluation
  5. Project closure

07. schedule

08

swot analysis

08. swot analysis

09

POTENTIAL OUTCOMES

CLOUD DATA

Data available in a cloud server to keep refining the model implemented into the user interface.

neural network-based interface

The expected outcome is a fully functional interface adapted to the Delta ASDA software, available to students during laboratory practices.

09. POTENTIAL OUTCOMES

10

references

A. IBM Cloud Education. (2022). Deep Learning. Retrieved November 20, 2022, from https://www.ibm.com/cloud/learn/deep-learning B. IBM Cloud Education. (2022). What is machine learning? IBM. Retrieved November 20, 2022, from https://www.ibm.com/cloud/learn/machine-learning C. Introduction to Neural Networks. (2020). Engineering Education (EngEd) Program | Section. Retrieved November 20, 2022, from https://www.section.io/engineering-education/introduction-to-neural-networks/ D. Zupan, J. (2003). Basics of artificial neural networks. Data Handling in Science and Technology, 199-229. https://doi.org/10.1016/s0922-3487(03)23007-0 E. Mitri, D. D., Limbu, B. & Klemke, R. (2021). Table Tennis Tutor: Forehand Strokes Classification Based on Multimodal Data and Neural Networks. Sensors, 21(9), 3121. https://doi.org/10.3390/s21093121 F. Plotz, T. & Guan, Y. (2018). Deep Learning for Human Activity Recognition in Mobile Computing. Computer, 51(5), 50-59. https://doi.org/10.1109/mc.2018.2381112

references

Next page

genially options