Posted in 2019

Diabetes detection using machine learning (part II)

In this 2nd post on detecting diabetes with the help of machine learning and using the Pima Indian diabetic database (PIDD) 1, we will dig into testing various classifiers and evaluating their performances. We will also examine the performance improvements by the data transformations explained in the previous post.

Read more ...

Diabetes detection using machine learning (part I)

Diabetes is one of the most serious health challenges today. The publicly available Pima Indian diabetic database (PIDD) have become a popular approach for testing the efficiency of machine learning algorithms 1. Within this context, this blog post is part of 2 posts providing an in depth introduction to diabetes detection using various machine learning approaches. In this first post in particular, we focus on exploring the data at hand and preparing it for machine learning related processing.

Read more ...

Voice based gender recognition using Gaussian mixture models

The aforementioned implementation, uses The Free ST American English Corpus data-set (SLR45), which is a free American English corpus by Surfingtech, containing utterances from 10 speakers (5 females and 5 males).

Read more ...

Authenticated encryption

With growing dependency on information technology and more at stake, the security aspect has become more vital than ever. Within this context acts authenticated encryption(AE) as a shared-key based transform whose goal is to provide secrecy, Integrity and authenticity of the encapsulated data 1. AE combines traditional Symmetric Encryption (SE) with a Message Authentication Code (MAC) in different orders 2. This article introduces the different combinations of authenticated encryption and their applications.

Read more ...

Encryption overview

Encryption transforms a plain-text into a cipher-text using an encryption key. The process of reversing this operation uses also a key and is called "decryption" 1. These operations can be explained by the following formulas: \(C=E_{k}(P)\) and \(P=D_{k}(C)\), where P = plain-text, C = cipher-text, E = the encryption method, D = the decryption method, and k = the key 1. There exist several types of encryption (symmetric/asymmetric) and several ciphering techniques (block/stream) and modes etc 1.

Read more ...

Cryptography overview

Cryptography is mostly associated with the development and creation of mathematical algorithms, which are used to insure various security aspects See 12. It is the cornerstone of modern communications security and is based on various mathematical concepts and theories such as: number theory, computational complexity theory and probability theory 3. The following post provides a quick overview of various cryptography concepts such as encryption, decryption and hashing.

Read more ...

Blog intro

This first blog is a quick introduction of my profile, and motives to start a blog. In this post I will be answering the following questions:

Read more ...