Week 13: November 20,2020
Short Quiz #3
1. Determine what kind of social media site you are creating.
For this assignment, I want to create a micro-blog for people who love to try new foods (home-cooked or from food chains/restaurants) or interact with other food lovers. Foodilicous will allow users to upload pictures of their food from different places and share their experiences with limited characters. Users may also connect with people through comments, likes, and followers. Users will also be able to retweet the food they like. The micro-blog will be divided into two sections â the for you page and the home page. The for you page will recommend different types of food not in the individualâs regular food palletâthe food categories in the for you page range from different cultures. The homepage will have content from the people they follow. I think this app could help usersâ widen their food pallet and learn about other cultures.Â
2. What will you measure?
The micro-blog will measure likes and retweets to showcase how many seem to enjoy it or how many users think the food is worth trying. I also plan to measure comments left under the blog posts because users will utilize words to share their experiences from the same restaurants or food chains.Â
3. How will you weigh these factors?
I will monitor comments left under the blog post for harassment or racists remarks. If comments do not meet the community guidelines, the app will do its best to not allow hate speech by replacing statements with asterisks and later removing it. Next, I will use likes and retweets to place recommendations on the for you page. Other usersâ engagement with a blog post determines the popularity and thus recommends new foods to try. However, every day there will be recent highlights from different food categories.Â
4. What factors will not be measured?
Foodilicous will not measure followers as it is not an influencer based app. The app focuses on food from different cultures. It will also not measure the usersâ gender, age, or religion to avoid cultural discrimination based on demographics.Â
5. How will the model learn or adapt to new information?
The model will learn or adapt to new information, like language, by adding a translation at the bottom of each post. If the userâs food pallet suddenly changes or adds something new, the model will have to relearn and adapt to recommend similar options.Â
6. What biases will be built into the model? How will this change the algorithmâs outcome?
One bias built into the model is receiving notifications for the food pallet users already have. For example, notifications from followers will showcase a preference for a particular type of food/culture. The top category of the for you page will be about food users are familiar with. The next categories will be from cultures users have not tried yet. The home page will also change the algorithm because users will only be exposed to food from their followers with similar taste buds. Thus, the user will limit what kind of content (food type) they will be exposed to.
7. How will the model live out the (un)ethical practices we have discussed?
I will prioritize the inclusivity of different cultures. For this app, it will not allow the typing of racial slurs or words. If a user attempted to write hate comments about peopleâs cultures, it will be replaced with asterisks and removed. Users can share their positive or negative experiences so long as it is not racially biased, sexist, or conveys any harassment. The app is heavily based on a participatory culture from its users. Thus, users are in control of who they want to interact with. They can block someone, such as a troll, and if many tend to block the same person, the app will ban the individual.Â










