No abstract

Machine learning is now becoming a widely used mechanism and applying it in certain sensitive fields like medical and financial data has only made things easier. Accurate Diagnosis of cancer is essential in treating it properly. Medical tests regarding cancer in recent times are quite expensive and not available in many parts of the world. CryptoNets, on the other hand, is an exhibit of the use of Neural-Networks over data encrypted with Homomorphic Encryption. This project demonstrates the use of Homomorphic Encryption for outsourcing neural-network predictions in case of Acute Lymphoid Leukemia (ALL). By using CryptoNets, the patients or doctors in need of the service can encrypt their data using Homomorphic Encryption and send only the encrypted message to the service provider (hospital or model owner). Since Homomorphic Encryptions allow the provider to operate on the data while it is encrypted, the provider can make predictions using a pre-trained Neural-Network while the data remains encrypted all throughout the process and finally sending the prediction to the user who can decrypt the results. During the process the service provider (hospital or the model owner) gains no knowledge about the data that was used or the result since everything is encrypted throughout the process. Our work proposes a Neural Network model which will be able to predict ALL-Acute Lymphoid Leukemia with approximate 80% accuracy using the C_NMC Challenge dataset. Prior to building our own model, we used the dataset and pre-process it using a different approach. We then ran on different machine learning and Neural Network models like VGG16, SVM, AlexNet, ResNet50 and compared the validation accuracies of these models with our own model which lastly gives better accuracy than the rest of the models used. We then use our own pretrained Neural Network to make predictions using CryptoNets. We were able to achieve an encrypted prediction of about 78% which is close to what we achieved when validating our own CNN model that has a validation accuracy of 80% for prediction of Acute Lymphoid Leukemia (ALL).

In this modern, technological age, the internet has been adopted by the masses. And with it, the danger of malicious attacks by cybercriminals have increased. These attacks are done via Malware, and have resulted in billions of dollars of financial damage. Which is why prevention of malware attacks has become an essential part of the battle against cybercrime. In recent years, Machine Learning has become an important tool in the field of Malware Detection, which is the first step towards removing malware from infected devices. In this thesis, we are applying machine learning algorithms to predict the malware infection rates of computers based on its features. We are using supervised machine learning algorithms and gradient boosting algorithms, such as LightGBM, Neural Networks, and Decision Tree Learning. We have collected a publicly available dataset, which was divided into two parts, one being the training set, and the other will be the testing set. After conducting four different experiments using the aforementioned algorithms, it has been discovered that LightGBM is the best model with an AUC Score of 0.73926.

<span>Embedded systems comprise several modules that exchange data by interacting among themselves. Exchanging wrong resource data among modules may lead to execution errors or anomalies. Interacting resources produce dependencies between two modules where any change of resources by one module affects the functionality of another module. Several investigations of the embedded system such as aerospace or automobile system show interaction faults between modules are one of the major cause of critical software failures. Therefore, interaction testing is an essential phase to reduce the interaction faults and minimize the risk. The direct and indirect interaction between modules generates interaction faults where indirect interaction is made underneath the interface in which data dependence relationship with resources may cause a different outcome. We investigate errors based on the indirect interaction between modules and introduce a new test criterion for finding errors detectable by existing approaches in unit level but not in integration level. In this paper, we propose a noble approach to generate an interaction model using indirect interaction pattern and design test criteria based on different interaction errors to generate test cases. Finally, we use fault injection and data flow coverage techniques to evaluate the feasibility and effectiveness of our approach</span>

No abstract