Instagram logs every interaction: likes, messages, stories viewed, searches, and follower changes. Even if you post nothing, your activity leaves a digital footprint.

In this talk, we will explore how to access and analyze the personal data Instagram collects about you. We will use Python, with libraries like pandas or polars, to clean and structure the information downloaded directly from the platform's privacy portal.

Next, we will build an interactive dashboard using Dash to visualize behavioral patterns over time. We will apply time-series anomaly detection techniques (such as rolling z-score, IsolationForest, or DBSCAN) and supervised classifiers (RandomForestClassifier, LogisticRegression) to identify potentially significant shifts in your digital activity—like a breakup.

Finally, we will reflect on the economic value of this data, the inferences platforms can make about our personal lives, and briefly discuss the European AI Act's relevance to privacy protection.