STATEMENT OF PURPOSE
Hi all
Any help would be greatly appreciated.
The motivation letter should be very concise and not longer than 1-2 pages. In general, we recommend a motivation letter of one page. Please do not repeat your resume in the motivation letter, but let us know why you would like to study in Germany, why you are applying to COMMAS, what makes you a good candidate and how you will benefit from the program and further information that will give us an impression about your aims and person.
My first entrance to the field of computer science was through a competition named Robocup which offered a free C++ course for the winners when I was 14. The course helped me understand the concept of programming and shaped my path in my future studies. Now with a deeper grasp of the field, I am intent on pursuing a master's degree in artificial intelligence and Data Science at the University of Stuttgart.
As I was doing my fundamental Java programming course in my first year of bachelor's, I came across Python and its ease of syntax in comparison to Java. I knew I had to delve deeper into the language so with the help of Al Sweigart's book, I was introduced to Automation using python. I was truly fascinated by how powerful python is and how much it can improve my life.
As a result of my newly-gained knowledge, I started integrating python into my watch-face startup. It was a game-changer for the business as it reduced our need for labor with the help of selenium and Beautifulsoup. By adding translator APIs we could sell our watch faces in 7 different languages corresponding to customers' needs. This is where competitors could not compete and as a result, our profits tripled. This made me realize I wanted to explore this field even further. Computer vision was truly challenging and even though I was working hard, I could only scratch the surface and build a script that can recognize the colors in a picture, find a corresponding pallet, and make the required watch face. But even to this extent, it was a breakthrough as we could make the design and the screenshots required by the Samsung platform completely automatic by using OpenCV and Pillow. After computer vision, I did a brief course on Udemy about NLP, where I learned how to recognize the emotions of a comment which was particularly important because we could know what type of watch faces had a generally positive overview and which did not without having to read the comments one by one. To visualize the results and to have a better command of what was happening, I learned to plot different charts using seaborn.
The next step seemed to be making an algorithm that could recognize which watch face would sell more even before it was produced, based on the background and the colors. Even though I put a lot of effort into learning how to use machine learning algorithms like random forest and K-means through titanic and Iris datasets. Knowing that as long as you are not able to make sense of the data you cannot find or make any algorithm that works led me to take a prerequisite Udemy course which covered a practical statistics tutorial using NumPy and pandas where I learned about the importance of Eigenvalue and Eigenvector in this field. It was not long before I realized for this particular job only deep learning and neural networks are of use. I tried finding my way through this problem by taking a course on convolutional networks, which gave me an understanding of the reward and punishment system work, but that was not the answer either. The more I learned, the more I realized how little I knew.
By the last year of my bachelor's degree, I was already deeply allured by Machine Learning and Data Science. Hence I picked deep learning and gender detection for my thesis which was ought to contribute to the development of Persian Pinterest. In the Lab, we worked on scrapping pictures from the web and labeling them using the Lableme library. Setting up Google Colab was not difficult as I had prior experience in working with Jupyter notebook. However, making the mistake of labeling YOLO instead of Pascal delayed my work significantly. Despite all the challenges involving incompatibility or compilation errors, I was able to overcome them through my self-learning abilities, as well as effectively seeking advice from others.
Upon completion of my bachelor's, I realized I needed further academic education to follow the passion I had developed for AI (especially NLP) and contribute to its development. The student startup I had was fun and it gave me educational profits alongside financial ones. But I realize I want to build something that helps humanity and solves problems and AI seems to be the answer. As a side project to my studies, I want to use my learnings in this course to ease the language learning process with the help of Machine learning and NLP alongside Stephen Krashen's method of learning languages with comprehensible inputs. So, if we can use AI to recognize our language knowledge gaps or run an analysis of our accents, fixing them will be less challenging. Nonetheless, none of these will be achievable without further education and professional guidance. And I know that Germany Is one of the leading countries in the AI and Data Science field and with Stuttgart being my favorite city, the cradle of the automobile, and the University of Stuttgart being among the bests worldwide, I am eager to do my master program in this program. The aforementioned master's program seems to be the perfect fit, because not only does it cover Language Processing, but also it helps me learn Human-Computer Interaction which is closely correlated with what I want to achieve. So, I think this is my opportunity to gain knowledge hoping I can give back to the community one day.