- Data Collection: Utilized vosonSML, rtweet, Rspotify, and tuber packages in Rstudio to collect all recent data about the artist through Twitter, Spotify, and YouTube API.
- Data Manipulation and Analysis: Leveraged tidyverse packages in Rstudio to manipulate the large amount of data collected from social media and extract insights about the artist's work, fan base, and other relevant information.
- Network Visualization: Created and visualized actor and semantic networks using Rstudio and Gephi to identify important actors and trending topics related to the artist.
- Data Visualization: Utilized ggplot package, adding layers, and customizations to visualize different audio features in every song, along with data of top trending videos on YouTube.
- Twitter User Clustering: Employed k-means to find clusters of Twitter users based on their friends and followers counts.
- Topic Modeling: Utilized Latent Dirichlet Allocation (LDA) for topic modeling of tweet text.
- Sentiment Analysis: Performed sentiment analysis on Twitter data using syuzhet package in R to classify tweets according to the sentiment and emotion.
- Buidling Decision Tree for ML Model: Used data from the Spotify API to train a decision tree model and evaluate its performance for predicting whether a song is by a certain artist, given the audio features of the song.
talia178 / artistdata_socialmedia_analytics Goto Github PK
View Code? Open in Web Editor NEWActing as a well-known artist’s/band’s manager, using social media analytics to help improve their popularity.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent