Link to our bot demo
Coronavirus makes us all stay at home and strikes fear in visiting public places, also confined with strict rules of social distancing. What happens with our moods if we can't go out? What happens to our bodies if we're stuck in an apartment for several days? What will happen after public spaces are open and free toall of us? How can we ensure safety of people of all risk profiles? We are not going back to normal. Our chat can save small businesses. Here is how.
What it does
When you want to go somewhere public, you address the chatbot and ask it for an suitable place in accordance to your risk profile. It recommends three ideal places for you given your location and your transport preferences within the 3 hours time period. After the visit you can provide the bot with feedback of the location you've visited.
How we built it
The first brainstorming session aimed to specify the goal, the challenge and to define work packages for the further procedure. It was also decided to use a chatbot as a solution. The rough draft for the construction was based on a typical user who needs assistance, as well as on a business owner in need of a customer. . The details that the chatbot queries for the persona were worked out in small groups. This includes a definition of the risk for yourself, the risk for others, questions about possible data sources, options for risk reduction through restaurants.
Challenges we concidered
The chatbot can be of great help to users in everyday life. Critical issues can be found above all in the areas of "data privacy" and "data security": on the one hand because of data storage, and on the other hand because the user should not disclose personal health data about himself. We solved the problem by not storing the data for more than 24 hours. There is also the question of how the chatbot is used by as many people as possible. Here, a balance must be accomplished between our technology and building upon the existing social media solutions.
Accomplishments that we're proud of
As an international team we worked well together and had a lot of group synergy as we were all excited about the challenge and possible solutions. Each team member contributed effectively with their own competencies and thus contributed to the overall success. Solutions to new challenges were quickly found together. We are very happy with the final product.
What we learned
The process of finding a common solution was very interesting. We got to know various new tools and methods for collaborating on projects. We also found that interdisciplinary teams with different competences can achieve excellent solutions to complex questions.
What's next for Eat Move Shop Bot
To go live with the chatbot we need to export the script to a tool like Dialogflow and acquire a business license to the WhatsApp API. In terms of datasources to match people to places, we'll need datasets about locations, like Google Business. Ideally we enrich the datasets with data such as surface area of a shop to derive the number of people allowed in the shop during the pandemic. For adoption of the chatbot, marketing is critical for success. The next step is the use and further development of good local databases. On the one hand, a Google API can be used, on the other hand, further data from local / national OpenData portals can also be used.
api, chatbot, google-places, whatsapp