TALA SIMO Cécile epse FOTSO
I am a Cameroonian, I defended the master 2 in December 2021 at the University of Ngaoundere and currently I filed my files for the thesis in this university. my objective is to analyze and improve the notion of responsibility in the solutions of cybersecurity using AI.
I am characterized by
passionate about cybersecurity and data science, i will like continous about this domaine. I love ravel,movies and music.
Professional experience
Organized online cybersecurity training, manager of the Moodle training platform; Community Growth Mutual , One-month academic internship in several positions in the company (MC2) in order to obtain the diploma of baccalaureate series: Information Technology.
Current responsibilities
Research Interests
My research field is cybersecurity, more precisely in artificial intelligence and scams.
Projects
Cogni-SHING
Level of studies
Ph.D
Current Studies
Ph.D ongoing
Description of current research
Advanced digital technologies and services, including artificial intelligence (“AI”) applied to specific tasks, hold tremendous potential. They have already spurred remarkable progress, including making many digital services more efficient, accurate, fast and easy to use. However, the emergence of these technologies has been accompanied by growing concerns about their possible adverse effects.more specifically in the field of cyber security. This implies not only better understanding the impact of technologies on solutions against cybersecurity attacks, but also closely studying the question of responsibilities in the event of harmful consequences, determining the aspects to be taken into account when design of such solutions and to propose a conceptual model to effectively apply responsible AI in cybersecurity.
Description of past research
Emotion recognition systems (ERS) are used in many areas of life, especially in the field of security to detect fraudulent conversations. In order to overcome this problem, we have proposed a model for the recognition of all the emotions during a telephone conversation in a general framework. A pre-processing step has been done on our audios in order to reduce the noise and to eliminate the silence in the set of audios. We applied two deep learning approaches namely LSTM and CNN to detect the emotions in the conversations in order to decide which approach would be best for our problem with good performance. We transformed our processed audios into spectrograms for our models. The proposed models were trained on these data, evaluated and tested to predict emotions in phone conversations.
Publications
Google scholar page
Social media pages
Personal/Business Web