Construct predictive model using the image analytics widgets

Reference no: EM133696966 , Length: word count:1000

Artificial Intelligence and Machine Learning

Assessment - Individual Evaluation Activity (Evaluating Neural Network Models)

Individual Implementation and Problem Solving (Puzzle)

Your Task

Evaluate the predictive modelling capability of neural networks and other prediction models in the Orange Data Mining Application. The assessment will involve implementing predictive models using software and solving various machine learning problems encountered in business.

Assessment Description
There has been a recent advent of Neural Networks including applications in deep learning. Analytics professionals can run basic Deep Learning applications via the browser and no-code platforms as the algorithms use hardware accessed via the cloud to provide the required performance.

In this assessment, you will implement an image classification model using Orange Data Mining's image analytics toolbox. You will be provided with a dataset containing images.

Assessment Instructions

In class: Implementation. Use the Orange data mining software to perform both deep learning- based image classification and classical predictive analytics tasks.

Image Classification -Construct a predictive model using the Image Analytics widgets in Orange.

Analyse tabular data using Neural Networks (NN)- Construct a predictive model using NN and other widgets in Orange.

In class: Problem Solving (Puzzle). Based on the Orange output, solve the accompanying puzzles in the assessment sheet designed to test your understanding of various prediction models and general knowledge on machine learning.

Here are the things you'd want to revise in preparation for assessment 2:

-Decision trees: prediction, class probabilities
-Perceptron model: prediction, training

-Neural network: image classification, embedding, network complexity

-Logistic regression: prediction, class probabilities
-Naive Bayes: prediction

-Word vectors: computing word vectors

-Orange workflow: data loading, train-test split, model training, testing

-Confusion matrix: matrix construction, computing precision, recall

Reference no: EM133696966

Questions Cloud

Compute the exponential smoothing forecast with alpha : Compute the exponential smoothing forecast with alpha=.2 and beta=.3 (you can use the results of first part and just add beta calculation).

Examples of external drivers of an organization risk culture : What are some examples of internal drivers of an organization's risk culture? Resource allocation, risk attitude, risk appetite, risk tolerance.

What are the types of malware being used today : What are the types of Malware being used today (Name two), and explain in your own words what they do.

Do you see any real-world applicability for coverage : Do you see any real-world applicability for Combinatorial coverage? And can you think of constraints that make it even more real-world?

Construct predictive model using the image analytics widgets : Construct a predictive model using the Image Analytics widgets in Orange and Analyse tabular data using Neural Networks (NN)- Construct a predictive model

Why is business ethics important : Why is business ethics important? What are some potential benefits to a business of having strong ethical standards?

What are some of challenges related to identity management : What are some of the challenges related to identity management? What are some of the IDaaS solutions offered by various companies?

Implement an image classification model using orange data : Implement an image classification model using Orange Data Mining's image analytics toolbox. You will be provided with a dataset containing images

Explain the goals that the model seeks to achieve : Explain the goals that the model seeks to achieve. Provide an overview of the IT and Cybersecurity departments' role in achieving Zero Trust.

User Account

All Pages