Is CitiBike an affordable and equitable transportation system with reduced fares for low-income residents, or is it a luxury service primarily benefiting wealthy New Yorkers?

Maheen Asghar

<https://github.com/miscalculation/citi-bike>

In this project, I analyzed data from CitiBike data from October 1st, 2019 to October 31st, 2019 (2,092,086 trips) to determine if CitiBike was an affordable system or if it was a luxury service. I found data to support the two opposing views.

![](/images/CitiBike_Map.png)


Using three different types of classifiers, we predict whether Trump and Obama's tweets are positive, neutral, or negative.

<https://github.com/miscalculation/sentiment-analysis-potus>

In the past decade, there has been much interest in the area of sentiment analysis, especially with regards to tweets. In our research, we focus on analyzing and predicting the sentiment of the past two presidents, Donald Trump and Barack Obama. Using three types of different classifier, we predict whether a tweet is positive, neutral, or negative, and evaluate our models using the F1 score and accuracy measures. The dataset used for training our models was taken from data.world and included tweets spanning from 2009 to 2016, when Trump was elected president. We use these predicted tweets to conduct deseasonalized time series analysis on the frequency of tweets for each president to explore specific time periods and the majority tweet sentiment for each president, as well their overall sentiment. We find that the Logistic Regression was the best-performing in both evaluation measures, for both presidents. Also, Trump’s tweets tend to be mostly positive before 2015, after which they become mainly neutral, while Obama’s tweets maintain their predominantly neutral sentiment throughout this time period.

![](/images/Obama_Trump.png)

![](/images/671%20Poster%20Final.jpg)


Sentiment Analysis of @POTUS

Visualization of data from the Opportunity Atlas.

<https://ai-income-vis.glitch.me/>

The Opportunity Atlas uses anonymous data following 20 million Americans from childhood to their mid-thirties. It gives data on how much kids who grew up in certain neighborhoods earn as adults.

Stanford University researchers found that almost half of parental income advantages in the United States are passed on to children. The research also found that a family's economic circumstances play an important role in determining a child's economic prospects. Children raised in families in the 90th percentile for household income can expect their own family income to be three times higher than children raised at the 10th percentile for household income.

According to the American Community Survey, 1 in 3 American Indians are living in poverty--with a median income of $23,000 a year. The findings at Stanford show that children raised in families that are "far apart on the income ladder can expect very different economic futures". Using the data from the Opportunity Atlas, we will be able to determine if the statement holds true with American Indians in the United States.

Click on the link above to read more.
