Hi, following is the abstract of my undergraduate thesis,if you’re interested, please contact me for further information.

[Abstract] Nowadays, E-business gets more and more popular in people’s daily life.During the online shopping, it would improve the shopping experience greatly if the users are allowed to search the intended goods by showing the online system with the pictures. This comes with the problem of large-scale content-based image search. In order to address this issue, one primary step is to collect sufficient pictures of varieties of goods for experimenting. In the thesis, the focus is to design a web-crawler to collect pictures from shopping website along with their meta data such as title, price, sales and URLs. In this work, 100,000 pictures have been crawled from taobao.com, which cover across varieties of goods, such as T-shirt, shoes etc. In the thesis, several latent issues related to crawler design have been addressed.