Customers within

How do you verify that you have created a new system that has the same functionality as the original? This is what Fredrik Öberg and his fellow student Magnus wanted to find out in their Bachelor's thesis

- We wanted to contribute something that could be used in practice. Together with Rasmus, who was our supervisor at YABS, we came up with a topic that would fit: formal specifications. We were also curious about the programming language Rust, which could be combined with this topic, says Fredrik.

The topic of the thesis (which lasted 10 weeks) was formal specifications, where the question of whether it is possible to use this tool to transfer functionality from the C programming language to the Rust programming language had to be answered. It was hoped that the use of formal specifications would increase the certainty that the transferred functionality would follow the original.

YABS connects and the app becomes a reality.

By chance, Lars-Göran (founder of Bizniz Tools) meets a Sylog consultant at an event in 2017. They start talking. The consultant then told them about their offering and specifically the subsidiary YABS that might be able to take on the task of further developing the BIZNIZ TOOLS app. Said and done. Lars-Göran and Lars put together a brief for how the app was supposed to work and in December 2018 the development work kicked off. From the start, three consultants and a consultant manager from YABS were involved in the assignment. Today, three years later, the app is available for download for iOS and Android. At YABS, a dedicated team is working on the continued development. The members of the team vary over time and many of the consultants at YABS have been involved in the development of the app.

This thesis has been done to automatically classify client reviews to determine if a review is positive or negative.

This was dealt with by machine learning. Three different deep neural network were tested on greater and lesser datasets, and compared with the help of two different frameworks, TensorFlow and Keras. Different embedding methods were tested on the neural networks. The best combination of a neural network, a framework and an embedding was the Convolutional Neural Network (CNN) which used the word embedding method Word2Vec, was written in Keras framework and gave an accuracy of approximately 88.87% with a deviation of approximately 0.4%. CNN scored a better result in all of the tests in comparison with the two other neural networks,Recurrent NeuralNetwork (RNN) and Convolutional Recurrent Neural Network (CRNN).