Projects

Dot

DOT is a personalized kids' activities booking platform. It uses recommendation algorithms and data science to personalize kids' activities booking based on kids’ interest, learning style, time, logistics, etc.

Transformative

STUDENTS INVOLVED

Lydia(Dan Liu)

ABOUT

Independent project

Help families find fun things to do, and make it as easy as checking your calendar

Dot aims to connect the needs of Millennial parents’ increasing demand for experiential consumption in China and the emergence of kids' activity start-ups, by effectively recommending personalized information to parents through the use of data and machine learning, and realizing a one-stop intelligent booking management system.

Millennial parents in China

Today, 90% of new parents are millennials. They spend a significant amount of time engaged in scheduled activities with their children. And there is so much to explore in more ways than ever before, that sometimes it’s hard for them to know where to start. They are the savviest of app users and the ones who spent on average of 5.7 hours per day using mobile devices. As millennials get married and have children, this group is more willing to invest money into experience consumption. China's "second child" policy has also promoted the development of the parent-child market. Data shows that the size of China's parent-child market reached 2.77 trillion RMB in 2018 and in the first-tier cities of China, parent-child consumption accounts for 60% of total consumption. Children’s activity sector is still highly fragmented, it is extremely difficult for parents to find and book the activities they want for their children.

Less time finding, more time enjoying

With Dot, parents don’t have to search through vast numbers of websites, forums and posts to find activities. Just download Dot and get started by telling us about your kids, including what their interests are, how they are doing in different activities and what is important when they choose a new activity. It’s as easy as messaging your friends. And in just seconds, they can get recommendations for each of their kids. Then, when they decide to book activities for their kids, they can just choose automatic mode, which is integrated into their calendar. Now they have all the information and don't have to keep going back to check their time availability. They can also manage everything in one place, like receiving tips or traffic reminders before the activities, showing QR codes to sign in, writing reviews or sharing with friends.

How does it work

So, what would a piece of recommendation technology driven by algorithm look like? Does it work in a real context? Here is a simplified picture of the recommendation algorithm data flow (shown on the right). By using a collaborative filtering and content-based recommendation algorithm, the users can get their initial recommendation through the user interface. Continuous analysis of the users' interactions (for example, their activity history, reviews, or even duration of spent on certain pages) sends feedback to the system, to continuously refine the recommendation.

Roadmap

To simplify the implementation process, the service can be developed in three different scales. Starting with an MVP, it can be implemented today with existing social media platforms like WeChat groups, which could be used as interaction interface and activity recommendation platforms. The service might lead to behaviour changing in a small group and certain areas. Then, the proposed service concept provides a mini app. It will enable data collection and some functions of the recommendation algorithm. The full ecosystem will enable us to shift attention to the organizers behind these activities and support both sides of Dot.

Projects

Dot

Less time finding, more time enjoying

How does it work

students involved on the project