Hello Everyone! This week our team was able to visit Tong Dun, 同盾科技, one of the data security firms that operate here in China (and other parts of the world too), whilst continuing to work on our theme project, diving deeper into constructing and testing the models that we decided on the previous week.
In the earlier portions of the week, we had our company visit to 同盾, where we learnt about some of their products and some of their future plans. They are a data security company that has products and solutions in a diverse range of industries. From Banking and Finance to E-commerce and even to Social Media and Gaming. They primarily focus on fraud detection but also offer services for privacy protection and risk management. They provide some of the back end solutions to online protection for both businesses and individuals. Their founder used to work at Alibaba as one of the leaders in the data security department before leaving to start up this company. As a relatively new company (around 15 years since starting up) the number of customers and solutions that they have provided as well as the number of industries that they serve currently is really something! Looking around the offices we saw much of the staff working in personalized work stations, and open spaces probably to encourage more collaboration between teams. It was an interesting experience being able to converse with one of the managers, who also happens to be an alumni of ZJU. Overall, a rewarding experience indeed.
The res of the week, we spent most of our time working on constructing and understanding the models that we have chose previously. Working with both EEG data as well as images, we experience quite a few hiccups. Being first timers when it comes to creating a machine learning model, we spent a lot of time trying to understand previously written code, data types, data structures and much of the other nitty gritty stuff that goes into creating a machine learning model. That being said, I think working together as a team we tend to see and fill in the gaps in our knowledge together and understanding the problems together, really reduced the amount of time spent if we had worked on it alone, collaborative learning really does work wonders. As we move forward, we’re hoping to be able to attempt to train these models over the coming week and hopefully improve on them so that by the end we’ll be able to move on to the next part of the project.
