Fhel Dimaano Email & Phone Number

Data scientist with a passion for finding the untold narrative in data. Experience in data acquisition, data visualization, statistical analysis, predictive modeling, machine learning, and natural language processing. With a background in psychology, psychiatric care, and customer service, I have worked with interdisciplinary teams across different functions and levels. I excel in client facing roles and communicating actionable insights to both stakeholders and clients.Worked both individually and in teams to develop and present a variety of data science projects including:• Predicting Positive or Negative Outcomes of 311 Requests in New York CityPredicting positive or negative outcomes of 311 requests using categorical models.+ Utilized NYC OpenData’s API to pull data on 311 service requests in New York City.+ Cleaned, feature engineered and used feature selection on data.+ Employed Logistic Regression, Decision Trees, Random Forest, KNN and XGBoost to determine which model returned the highest metric for making predictions using F1 Scores.• CNN Facial Classifications: Emotion, Gender, RaceA convolutional neural network that classifies emotions, race and gender.+ Trained convolutional neural networks with over 45,000 images and 16 classes.+ Utilized Keras to build convolutional neural networks to classify emotion, race, and gender in images.+ Applied Transfer Learning using VGG16, InceptionV3 and Xception models.• Whiskey Recommendation SystemA model based recommender system to expand a user’s whiskey palette.+ Created a model based recommendation system based on a user’s budget and ratings.+ Cleaned and manipulated data consisting of 41,000 reviews, 1,200 unique users, and over 1,000 unique whiskies.+ Employed Singular Value Decomposition, K Nearest Neighbors and GridSearch to determine which yielded the lowest Root Mean Squared Error as the metric for prediction.

Listed skills include Artificial Neural Networks, Neural Networks, Scikit Learn, Pandas, and 24 others.