Cornell Task: 2016' Peiyi Shen

Peiyi Shen
Professor Jason Hogg

【Reflection of JD Hackathon】
The hackathon is very impressive, the students from our program and the students from Cornell tech were regrouped, it means that each team has students from china and oversea. The theme of the hackathon was given by the sponsor JD.com, though team work we should give a presentation about how to help JD.com in improving in intelligent supply chain, retail intelligence and dynamic pricing.

During the two days, we should collaborate to come up with the best solution and prepare the final competition. Students from different country have their own knowledge and cultural background, so cooperation could crash out many wonderful ideas, but it was also because of this, the communication barriers may also create a lot of different opinions. Since time was limited, we still want to make more innovation in technique and business model.

Our team had a few discussions before the event and we had chosen the retail intelligence as our topic, we researched and analyze the current situation of JD.com,form our Tsinghua-Cornell students’ prospection, we all recognized that JD.com’s commodity system is not perfect, if we can do some valuable innovation on comments sold on the JD.com, then can improve the user’s residence time on the page, the purchase conversion rate can be obviously improved. Many researches and studies showed that the recommendations from the people who we trust will help you a lot when we were making decision about whether to buy the goods or not. Therefore, we hope to be able to grab the information about the goods from social media including the relevant images, text, video and other information, combined these information with the semantic analysis and emotional analysis methods, finally to show these information aggregation websites about the goods then the comments may be more valuable than nowadays. The Cornell Tech student chose another aspect about retailing.They want to do something to improve the sales of clothing category,JD’s clothing retail business is weak compare with its biggest competitor Tmall, their idea is to establish a fashion lab project for JD.com, they will collect customer information on the be fond of clothes from offline store, and analyze the information such as the trend of popular magazines, finally get the popular trend and feedback to clothing manufacturers, expecting to make a more suitable for the requirements of the dress. When we discuss the pro and con of the two ideas, we realized that the respective opinion are based on different perspectives, we knew that JD was one of China's most successful enterprises, everybody is JD.com’s client, we think we know more about what they need and their future development strategy, our scheme is more like realistic situation, even can bring real value for JD.com, but oversea students think more powerful and unconstrained style, creative, able to attract the attention of people. So the best solution is actually a fusion of both advantages, formed a comprehensive solution. After a long discussion, we decided to adopt the idea of the Cornell Tech students, but in one of the key part can use our model, we set up a evaluation system,which could regression to all kinds of factors that affect the clothing fashion, used to predict the trend, further improve the efficiency of the sales of intelligent degree and in the supply chain.

Eventually our plan didn't get the top three position, I'm sure that we have many shortcomings, in terms of the cultural differences is one of the most important and improve the efficiency of communication is the most appropriate way.