Customers need personalized meal suggestions that are both in line with their health goals and their food tastes. No matter if the customer is on low sodium, low-cholesterol or on a fully personalized DNA profile-driven diet they can get the right suggestions. Is your customer in need of only recipes that are vegetarian, contain less than 600mg of sodium per serving, are high on Vitamin A but low in fiber? PHILIA Alexa Conversations Skills can help with this.
What it does
PHILIA is an Alexa Conversations Skill, that provides over 2 million recipes that are indexed, normalized, and contain full nutrition information. The skill allows for search by nutrient quantity, 40 diet, and health labels as well as keyword searches by cuisine type (chinese, italian, indian, french,...), meal type (lunch, dinner, breakfast, snack) or dish type (soup, salad, pizza, sandwich,...).
How I built it
It is not a basic or easy skill, as it is developed with innovative technologies such as:
- Amazon Alexa, Amazon’s cloud-based voice service available on hundreds of millions of devices from Amazon and third-party device manufacturers.
- Edaman, Recipe Search API, Food Database API, and Nutrition Analysis API
- Python3.7 , programming language.
- AWS Route 53, a highly available and scalable cloud Domain Name System (DNS) web service.
- AWS Lambda, a computing service that runs code in response to events and automatically manages the computing resources required by that code.
- AWS CloudFront , Fast, highly secure and programmable content delivery network (CDN)
- AWS api-gateway, Create, maintain, and secure APIs at any scale
Challenges I ran into
Every task / MVP brings new challenges, tests the efficiency of individuals and teams. To complete any MVP successfully it requires a lot of effort, hard worker. Along with technical/development skills, task completion also needs a lot of other team-based skills. It tests an individual's ability, team coordination, communication skills.
It’s a very difficult task to recommend food mainly based on their nutrients requirement person like because different people like different types of dishes . It varies according to their origin place, their current living state. Also person the budget also matters a lot for what kind of food he prefers instead of nutrition. But we optimized our result to get much-related recommendations.
Accomplishments that I'm proud of
We are proud to say that finally, we complete our task successfully with decent results as the saying goes "hard work pays off". This is the complete technology for the whole time we keep on updating. Its MVP entire, lots of additional features will come soon. we keep updating this with better results.
What I learned
We learned a lot of development as well as team-based skills:-
- Voice design dialog
- Building a skill using Alexa Conversation
- Training Alexa Conversations to collect informations
- Learn about lots of APIs which I don't know before like Recipe Search API, Food Database API, and Nutrition Analysis API,
- How to develop a recommendation system.
- How to avoid overfitting in machine learning models.
- Learn a lot of team-based skills how to break a complex task into lots of parts and steps
- Refine understanding challenges through discussion and explanation
- Time management, Team leadership quality the most important skill needed to complete any task, Good communication with teammates
- How to apply development skills in the real-world in a beneficial of the world
What's next for PHILIA
We plan to enhance the accuracy to make it more user friendly by using the following steps:-
- Try to merge best-suited recommendations based on origins and amount of nutrition diet recommendation system to get optimized results which customer like.
- increase the variety of dishes and user-specific recommendations.
Currently, we are covering certain regions' food data in recommendation an MVP so we are covering a small region but soon we try to increase a wide variety of foods to cover whore world in our MVP.