Clouies-- Investment Assessing System Based On Big Data

Project Icon

Inspiration

Our idea comes from the growing lower effeciency digging data for a certain company when making investment. As the big data and artificial intelligence is more and more neglectable in life, we are trying to combine it with the investment to save more energy for searching and processing data.

What it does

The project aims at helping investment company or banks making better decisions when making investment. It generalize a score for the company which the user searching on our project. The project returns the final score from 0 to 10 , a spider-web-plot and detailed scores in five aspects:finance,environment,developing,security, and core competitiveness. Thus, the invester can make a wiser decision.

How we built it

It fetches data of the company from the website. Then the project will process the data, search the key words in every record, group the message into five categories, and finally grade the message influence towards the company. When generalizing the final score of the company, we add the score from the five categories with certain ratio and report it to the user with details.

Challenges we ran into

-It is extremely difficult to test and train the model with essays or news title from the internet. -The spider to get data is hard. -Getting in trouble with generating the list of key words.

Accomplishments that we're proud of

-We finally realize getting data from Twitter and GoogleNews. -The model for scoring the title is working in a resonable way. -We make a visiable data preview in the result UI.

What we learned

We learn a lot knowledge about spider in Python and searching key words which we never use in the past. We learned the importance of business prototype and business model for a project. We learned how to work together and combine our ideas together.

What's next for Clouies-- Investment Assessing System Based On Big Data

For now, we only have the data from Twitter and Google News. For next, we will expand the list of our data source. What's more, we will train the AI based on our five aspects. If possible, we will make the semantic recognition for the whole essay or news which can make the score more accurate and resonable. Finally, we can improve the interface and providing the user with a better experience.