Finvisor

The project is based on the idea that it will take the user financial data, analyse it, and predict its view on the user future Savings and possibilities of investments based on the given dataset.

Comment

Inspiration

We are a team of four First time in in-person hackathon and we came here to learn as much as possible. None of us has in depth knowledge of backend technology. However, Tatra Banka has articulated the challenge in a very simplistic way and also their workshop on Figma helped us to build the prototype.

What it does

We could not build a functional working application. However, we have designed a prototype using Figma which demonstrate the user friendly flow from an individual that has no knowledge of technology.

  1. At first the app will take the user's name and age.
  2. Then it will navigate the user to the pages where it will collect income and expenses
  3. Then user will be able to add multiple month's data.
  4. User will be able to generate savings prediction for upcoming month.
  5. If user wants to get prediction about their investments, they will need Sign up(Monitised)
  6. After signing up, user will need to login and then they will be able to use the Investor prediction feature.
  7. The investor prediction is a tailored recommender system base on the risk profiles, budget, debt, expenses, history and goals.

How we built it

As we only have basic of python and not knowing the knowledge og plugin integration of Figma and Python, we have tried writing codes getting various sources from online.

For example below code will generate dataset (as personal finance data are not publicly available)/ dates = pd.date_range('2020-01-01', '2021-12-01', freq='MS') income = np.random.normal(loc=5000, scale=1000, size=len(dates)) rent = np.random.normal(loc=1000, scale=200, size=len(dates)) utilities = np.random.normal(loc=300, scale=50, size=len(dates)) groceries = np.random.normal(loc=500, scale=100, size=len(dates)) transportation = np.random.normal(loc=200, scale=50, size=len(dates)) otherExpenses = np.random.normal(loc=400, scale=100, size=len(dates))

data = pd.DataFrame({'Date': dates, 'Total Income': income, 'Rent Expense': rent, 'Utility Expense': utilities, 'Grocery Expense': groceries, 'transportation': transportation, 'entertainment': entertainment, 'otherExpenses': otherExpenses}) data.set_index('Date', inplace=True)

Challenges we ran into

The biggest challenge we ran into is to development and integration of all the codes we have gathered. Simply we did not have any knowledge. We have tried adding multiple plugins and watched a lot of videos but not entire E2E process found that could help us.

Accomplishments that we're proud of

We have learned so much, None of us touched Figma before, and now we have a cool smart navigated prototype that we are really proud of.

What we learned

We have learned a lot of bits and pieces of data set tests and how to use basic of numpy, tkinter and panda

What's next for FinVisor

We think the biggest step will be give our Figma smart screens a life with Python integration, implementation and possible create/generation of a larger dataset to train and test for the recommender system.

Built With

Share this project:

Updates