This report investigates how musical attributes and structural metadata influence song popularity on Spotify. We explore whether attributes such as danceability, energy, loudness, acousticness, speechiness, tempo, and valence, along with key, mode, time signature, and genre, interact to shape listener engagement and ultimately drive a song’s popularity. The analysis is structured into several key sections: - Data Preparation & Cleaning: Importing and cleaning the dataset. - Exploratory Data Analysis (EDA): Visualizing relationships and distributions of variables. - Statistical Modeling: Using regression analysis to predict song popularity and checking the model’s assumptions. - Clustering Analysis: Grouping songs based on their attributes to identify patterns. - Conclusion & Recommendations: Summarizing the insights and suggesting next steps. Each section includes detailed explanations so that even readers new to the subject can follow the rationale, methods, and results.

The European Social Survey (ESS) provides high-quality open access data measuring public attitudes, beliefs and behaviour. In today’s paper I’ll be inspecting the whole dataset and the special variable “hincfel” feeling about households income nowadays.These results provide insight into economic well-being across European populations, highlighting disparities in perceived financial security. I’ll use only Apriori algorithm to uncover some patters, keeping “hincfel” as a rule consequent. Later, there will be appropriate vizualizations such as parallel coordinates plot, bar chart, scatter plot, network graph and lastly sankey diagram.