GETPREG

Inspiration

We are Getpreg team. We aim to serve the needs of infertility patients via an online platform. Infertility patients are a group of people who are difficult to get pregnant. Although lots of marriages in a year, only a few of them are successful in getting pregnant. One of the reasons behind this is that they don’t know the exact problem and actual solution suited. They seek tons of advice from many experts and need support from everyone. So, we decided to create a new online consultant platform that will help them reach out to the experts or infertility clinics easier and help them plan and prepare for becoming parents. We want to deliver them the most efficient system that provides initial advice and recommendations from the personalized recommendation system.

What it does

A personalized recommendation system will be very crucial for our project, as we need to collect any personal data from the clients. The system will pre-screen and classify whether the customer is infertile or not based on their conditions. We aim to build the automatically respond system and give personalized recommendation plan that suits their conditions. Moreover, providing the best experiences for our customers. The system will communicate with customers naturally and analyze the data from the input to suggest the best option that suits their needs and concern.

How we built it

Generally, we use messaging API from Line as a receiver collecting all data needed to be processed. then, it is linked to Dialogflow to generate a conversation flow, training the phrases and being able to answer that is not exactly the same as training inputs. next, we also access the past database, in this case, is the history treatment. lastly, it will be returned to the line OA to give the user the answer. So our architecture involves Line, Dialogflow, and database in order to generate new recommendations

Challenges we ran into

The first challenge that we ran into is that the limitation of scope and flexibility of responses since the information will be based only on the programmed data server. The next challenge is that the understanding of the complex human context. Lastly, is the accuracy and legibility of responses provided by the system.

On another side of the work, we still struggle with the coding part we have not integrated that into our project, yet we will try in the near future.

Accomplishments that we're proud of

Business models are fundamentally linked with technological innovation. After we explore more on the AI field, our business model has changed to a better way. A personalized recommendation system can serve our customer's needs since they are struggling with gathering a variety of information and making a decision. The system adds value to our model and differentiates us from our competitors.

What we learned

As our team wants to create a personalized consultation service for infertility patients as well as providing the best and efficient suggestion for our users. We decided to conduct the communication channel via message API line. According to this, we learned how to create a Human-like conversation generated by Dialogflow where the fundamental knowledge of building software architecture design and development are also required. Most importantly, we received the skills of applying the implementation of artificial intelligence and technology business to reduce the complexity of the data as well as reduce time-consuming tasks.

What's next for GETPREG

We want to increase the accuracy rate of the algorithm by backtesting the bot with the past cases. This will ensure that our bot will provide the best match suggestion for each case. Next is developing a horoscope bot or partnering with an online fortune-telling company as mostly Thais people believed in lucky dates and numbers. Infertility treatment is also emotional to the patients, so we find this as a business opportunity where we can enter. After gaining patients, creating a marketplace for motherhood and childhood products is also what we aim to do.

Team members and Roles

1.Thanapat Phatraachariyakul 6258027356 (Conversation flow and a bit on dialog flow) 2.Thanaree Thongmee 6258028056 (System flow) 3.Nathnicha Surattanavongkul 6258020956 (Data gathering) 4.Saruta Kreangpichitchai 6258069756 (Dialog flow)