Reference no: EM133779539
Question: In the old days, optical character recognition (OCR) systems worked by comparing the scanned image to one of the images previously stored images, and the hit-and-miss rate to these earlier files defines the recognition efficiency. Currently, OCR uses a set of AI techniques for character recognition, including neural networks. For instance, neural network technology is used to analyze different things, including the stroke edge, the line of discontinuity between the text characters, and the background. OCR might use different AI techniques and average or poll the results out of all of them.
One of the most used datasets is the MNIST (LeCun et al., n.d.). This dataset is used as a base for measuring the performance of any developed OCR algorithm. It is used for Kaggle competitions as well.
In this Assignment, we will be dealing with the OCR problem. You are required to develop a solution to a handwritten character recognition problem using artificial neural networks and deep learning algorithms. There are different performance measures; you are required to report the following:
Precision: The number of true positives/(the number of true positives + false positives).
Recall: The number of true positives/(the number of true positives + the number of false negatives.
F1 Score: Weighted average of precision and recall.
To complete the Lab:
Perform the following tasks:
Develop a neural network algorithm to identify the handwritten characters.
The algorithm could have a minimum one input layer, one hidden layer, and one output layer.
Implement the developed algorithm using any language of your choice such as Python, Java, C/C++, R and MATLAB.
Note: the MNIST dataset is formatted to be used with MATLAB; however, it could be used with any other programming environment/platform such as TensorFlow, Neuroph, and Deeplearning4j.
Record the recognition performance (precision, recall, and F1 score) for the following cases:
Using 100, 200, and 500 neurons in the hidden layer, using the MNIST datasets.
Use at least two activation functions.
Examine the handwritten character recognition problem, using one of the deep learning algorithms.
Observe the developed deep learning algorithm performance (precision, recall, and F1 score) on the MNIST dataset.
In a 5- to 6-page paper in APA format, provide a detailed description of:
Both developed algorithms
The used dataset
The performance of neural network algorithms
Your observations
Screenshots
- Access the Wireshark sample pcap files
- You will also need to use Git or PuTTY.
Submit a 3- to 5-page summary report in APA format in which you:
- Explain the steps you took to complete the lab. Include screenshots for the major steps.
- Identify the pcap file chosen from the Wireshark sample captures website and detail the Snort rules you wrote to detect the intrusions recorded in the file. Include screenshots.
- Detail the Snort outputs of your Snort rules on the pcap file. Include screenshots.
Your document should be 3-5 pages long (not including the list of references), but it is the quality of the work that is important, not the number of pages. Cite and reference all sources using APA format and style guidelines and submit in a single document.
|
Create a program implementing a given real-world problem
: Create a program implementing a given real-world problem. You will design your databases, use Python to link to a central database, and perform the application
|
|
Create well-crafted chronological or functional resume
: Create a well-crafted chronological or functional resume,that showcases your academic and professional achievements,
|
|
Increase its productivity as measured
: A boot factory wants to increase its productivity as measured by the number of boots that are processed per hour at each of its facilities.
|
|
HR data regarding employee demographics
: An advertising agency has collected HR data regarding employee demographics, performance ratings, years of service, etc.
|
|
Wrote to detect the intrusions recorded in the file
: MCYB 6670 Security Engineering and Compliance, Walden University - Explain the steps you took to complete the lab. Include screenshots for the major steps
|
|
Synthesis of prior research and the application
: Synthesis of prior research and the application of ethical principles and multicultural assessment practices
|
|
Issues addressing acquisitions
: Read a journal article on issues addressing acquisitions. Summarize the article. Article needs to address the overall PPBE process
|
|
Total quality management approach
: Despite not using a Total Quality Management approach, my organization has used a different approach called Quality Improvement Plans
|
|
Prepare forecast for demand for financial advisers
: As part of its internal analysis, an HR manager at a financial services firm is asked to prepare a forecast for the demand for financial advisers
|