ASSIGNMENT-II

Q1. A case study on Application of Neural network in any one of the fields
a) Biomedical Imaging
b) Satellite Imaging
c) Financial prediction
d) Big data Application
e) Cyber security
f) Image fusion

Introduction

Here we can give a consideration to a problem of detecting errors and vulnerabilities

in software components of different digital devices. The details shows an ever
increasing criticality of this problem in the course of time related to development of
modern concepts the Industrial Internet and the Industry 4.0. It gives an overview of
modern approaches to application of methods of computer-assisted learning and
artificial intellect in the sphere of cyber security, problems and prospects of
application thereof. A new approach is offered to searching software vulnerabilities
on the basis of application of deep learning. The approach is based on building
semantically significant vector representations of software code and multistage
instructing the deep neural network on revealing hierarchical abstractions in computer
code testifying to presence of vulnerabilities. The authors describe specific features of
the goal of analyzing software code for presence of vulnerabilities and proceeding
there of it is offered to use a neural network with long short-term memory (LSTM). In
order to solve a problem of the learning set, the authors offer to use learning with
transfer in case of building vector representations of instructions. This detail also
provides results of experimental investigations on application of offered solutions.

Approach

Neuro-semantic approach to searching vulnerabilities

1) Analysis of high-abstract program characteristics
2) Building vector representations of program instructions

Conclusion

The development and implementation of this approach will help develop solutions
providing automatic detection of vulnerabilities in the program components of
different digital devices over acceptable time without using exponential memory
consumption. This is especially relevant under conditions of dashing increase of the
number of digital devices accessible through Internet. The incorporation of this
 approach in practice will help, if necessary, to additionally learn the network with
new examples of the vulnerable code (with new types of vulnerabilities). Besides, a
modular architecture of computation architecture provides for a possibility of
supporting different hardware architectures of processors in the way of learning
vector representations for different sets of instructions. These results can be used for
advanced detection of and removal of program vulnerabilities in the code of different
digital devices being protected.

Q2. Ensemble NN Applications

An ensemble neural network (ENN) technique has been proposed for PD pattern
recognition. The results of the ENN have been compared with the widely applied
single neural network (SNN). In recognizing PD patterns of the same training and
testing data, the ENN classifier has an improved classification performance over the
SNN.In recognizing PD patterns different from training data, improved accuracy and
performance is still shown by the ENN. For practical applications, the ENN can be
considered as a potential PD pattern recognition tool.

Q2. Tackling COVID -19 with NN and CNN or Deep Neural networks

COVID-19 Open Research Dataset, which provides the text of over 24,000 research papers,

the COVID-net open-access neural network, 

Additional Resources

 COVID-19 Research Database (provided by the WHO)

 LitCOVID (provided by the NIH)
 COVID-19 Resource Page and CORD-19 AI Powered Search (provided by
Microsoft)
 COVID-19 Research Export File (provided by Dimensions)
 Day-Level COVID-19 Dataset (hosted on Kaggle)
 COVID-19 Global Cases (provided by Johns Hopkins University)
 COVID-19 Open Patent Dataset (hosted by Lens.org)
 COVID-19 Global Rheumatology Registry (provided by the Global Rheumatology
Alliance)
 COVID-19 Literature Review Collection (hosted by Cochrane Library)
 CORD-19 Search Engine (provided by Verizon Media)
 COVID-19 Policy Dataset (provided by Overton)
 COVID-19 DOC Search Engine and Resource Page (provided by Doctor Evidence)
 COVID-19 Clinical Trials Tracker (provided by TranspariMED)
 ASReview - Active learning for Systematic Reviews (provided by Utrecht University)
 Blog Post: Computer Scientists Are Building Algorithms to Tackle COVID-19