what

Assignment VI/DataAnalytics2025Fall_A6_Itamar_Oren-Naftalovich.md

@@ -0,0 +1,410 @@
+# News Popularity in Multiple Social Media Platforms
+
+This project analyzes the **News Popularity in Multiple Social Media Platforms** dataset from the UCI Machine Learning Repository. The data contains ~93k news items collected between November 2015 and July 2016, with their final popularity on Facebook, Google+ and LinkedIn across four topics: *economy*, *microsoft*, *obama* and *palestine*.
+
+---
+
+## 1. Exploratory Data Analysis
+
+### 1.1 Data overview and cleaning
+
+We work primarily with `Data/News_Final.csv`, which has **93,239** rows and 11 variables:
+
+- `IDLink` – numeric id of the article  
+- `Title`, `Headline` – short text fields  
+- `Source` – news outlet that originally published the story  
+- `Topic` – one of {economy, microsoft, obama, palestine}  
+- `PublishDate` – publication timestamp  
+- `SentimentTitle`, `SentimentHeadline` – numeric sentiment scores derived from title and headline text  
+- `Facebook`, `GooglePlus`, `LinkedIn` – final popularity on each social media platform
+
+According to the dataset documentation, **-1** in the popularity variables indicates that no final popularity value was observed. In the code, any value `< 0` in `Facebook`, `GooglePlus`, or `LinkedIn` is therefore replaced with `NaN`. Missing popularity values are later dropped on a per–model basis. `PublishDate` is converted to a proper timestamp, and a numeric time feature
+
+```text
+DaysSinceEpoch = days since 1970-01-01
+````
+
+is created to allow inclusion of temporal trends in the models. We also log‑transform Facebook popularity:
+
+```text
+log_Facebook = log1p(Facebook)
+```
+
+which is used as the target for regression models.
+
+---
+
+### 1.2 Popularity distributions
+
+A histogram of Facebook share counts on a **logarithmic x‑axis**, after removing missing and zero values
+
+![distribution of facebook popularity](imgs/eda_facebook_hist.png)
+
+*Figure 1: Distribution of Facebook popularity on a log x‑axis.*
+
+The distribution is extremely right‑skewed:
+
+- Most articles receive very few shares.
+- A small number of “viral” articles receive thousands of shares.
+
+On the cleaned data, summary statistics for Facebook shares are approximately:
+
+- median is approx 8
+- mean is approx 129
+- 90th percentile is approx 214
+- 99th percentile is approx 2,322
+- max = 49,211
+
+Google+ and LinkedIn exhibit similar heavy‑tailed patterns (with smaller absolute scales), which matches the description of the dataset creators ([arXiv][1]).
+
+The distribution of `log1p(Facebook)`
+
+![distribution of log-transformed facebook popularity](imgs/eda_log_facebook_hist.png)
+
+*Figure 2: Distribution of log‑transformed Facebook popularity.*
+
+The log transform compresses the heavy tail and produces a more regular, unimodal distribution. This justifies using `log1p(popularity)` as the regression target: it reduces the influence of rare extreme outliers while keeping them in the data, which is important because viral stories are the phenomena of interest.
+
+---
+
+### 1.3 Topic effects
+
+The four topics are not equally represented:
+
+- economy: 33,928 items
+- obama: 28,610
+- microsoft: 21,858
+- palestine: 8,843
+
+The mean log‑Facebook popularity by topic.
+
+![average facebook popularity by topic](imgs/eda_mean_by_topic.png)
+
+*Figure 3: Mean log‑Facebook popularity by topic.*
+
+Key observations:
+
+- **obama** stories clearly have the highest average popularity.
+- **microsoft** is slightly above **economy** and **palestine**.
+- In original share counts, obama articles average roughly an order of magnitude more shares than economy/microsoft/palestine stories, but all topics remain strongly skewed.
+
+This suggests that topic is an important categorical predictor for popularity, and motivates including it as a one‑hot encoded feature in the models.
+
+---
+
+### 1.4 Sentiment and popularity
+
+Sentiment scores from the title and headline are continuous values roughly in the interval [-1, 1]. Their empirical distributions are centered very close to 0 with standard deviations around 0.14, indicating that most titles and headlines are only mildly positive or negative.
+
+A 5,000‑row sample of `SentimentTitle` vs `log_Facebook`
+
+![title sentiment vs facebook popularity](imgs/eda_sentiment_vs_popularity.png)
+
+*Figure 4: Scatter of title sentiment vs log‑Facebook popularity (sample of 5,000 articles).*
+
+The scatter plot shows:
+
+- A dense vertical band near sentiment 0, reflecting many neutral titles.
+- Viral and non‑viral articles scattered across the full sentiment range, with no obvious linear trend.
+
+Empirically, the correlation between sentiment and Facebook popularity is almost zero (|r| is approx 0.01). This suggests that sentiment alone is a weak predictor of popularity; we still include it in models because it may interact with topic or time, but we do not expect it to explain much variance by itself.
+
+---
+
+### 1.5 EDA conclusions
+
+From the exploratory analysis we conclude:
+
+1. **Popularity variables are non‑negative, highly skewed, and heavy‑tailed.**
+
+   - Log‑transforming shares yields more regular distributions, so regression models should target `log1p(popularity)` instead of raw counts.
+
+2. **Topic has a strong effect on expected popularity.**
+
+   - Particularly, obama‑related news is more popular on Facebook; microsoft is relatively stronger on LinkedIn (from descriptive statistics, not shown here).
+
+3. **Title/headline sentiment has little linear relationship with popularity.**
+
+   - It should not be expected to drive predictions strongly.
+
+4. **There are many extreme outliers (viral stories), but these are the signal we care about.**
+
+   - We choose *not* to remove them; instead, we rely on robust models and log‑transformed targets.
+
+These observations motivate a modeling strategy that combines:
+
+- **Linear models** (to quantify simple topic/sentiment effects on log‑popularity).
+- **Non‑linear tree‑based models** (to capture complex relationships and heavy‑tailed behaviour).
+- **Classification** of viral vs non‑viral stories.
+- **Clustering** of time‑series trajectories to identify typical growth patterns.
+
+The next section formalizes these ideas.
+
+---
+
+## 2. Model Development, Validation and Optimization
+
+We develop **five** models: three regression models (including a dimension‑reduced variant), one classification model, and one clustering model. This covers regression, classification and unsupervised learning objectives, and explicitly examines the impact of dimensionality reduction.
+
+All supervised models use:
+
+- Train/test split: **80% training, 20% test**, `random_state=42`.
+- Evaluation on the held‑out test set only (no peeking).
+- Metrics:
+
+  - Regression: R² and RMSE on log‑scale (using `root_mean_squared_error`).
+  - Classification: accuracy, F1 for the positive class, ROC AUC and confusion matrix.
+
+### 2.1 Common preprocessing
+
+For each model:
+
+1. Replace `-1` in `Facebook`, `GooglePlus`, `LinkedIn` with `NaN`.
+2. Drop rows with missing values in the specific target variable.
+3. Use `DaysSinceEpoch` as a numeric representation of `PublishDate`.
+4. Where appropriate, use `log_Facebook = log1p(Facebook)` as the regression target.
+5. Encode `Topic` using one‑hot encoding with economy as the reference level (`drop_first=True`).
+
+For time‑series models we also use `Data/Facebook_Economy.csv`, which stores Facebook popularity snapshots TS1–TS144 every 20 minutes for economy articles. We join it with `News_Final.csv` on `IDLink` and restrict to:
+
+- `Topic == "economy"`
+- Time slices **TS1–TS50** as predictors (roughly first 16–17 hours)
+- Final log‑Facebook popularity as the target
+
+Negative TS values are interpreted as “no observed popularity yet” and are set to 0.
+
+---
+
+### 2.2 Regression Model 1 – Linear regression on static features
+
+**Goal.** Predict log‑Facebook popularity using only static metadata (no early popularity feedback).
+
+- **Target:** `y = log_Facebook` for all topics.
+- **Features:**
+
+  - `SentimentTitle`, `SentimentHeadline`
+  - `DaysSinceEpoch` (publication time)
+  - Topic one‑hot dummies: `Topic_microsoft`, `Topic_obama`, `Topic_palestine` (economy is implicit baseline).
+
+We fit an ordinary least squares linear regression on the training split and evaluate on the test set.
+
+**Results (test set):**
+
+- **R² is approx 0.157**
+- **RMSE is approx 1.86** in log‑space
+
+Actual vs predicted log‑Facebook values
+
+![model 1: actual vs predicted](imgs/model1_actual_vs_predicted.png)
+
+*Figure 5: Model 1 predictions vs actual log‑Facebook values.*
+
+The predictions are compressed into a narrow band, under‑predicting viral articles and over‑predicting low‑popularity ones. Key coefficients:
+
+- `Topic_obama` is approx +1.78 (large positive shift vs economy)
+- `Topic_microsoft` is approx +0.10
+- `Topic_palestine` is approx +0.02
+- `SentimentTitle` is approx −0.38, `SentimentHeadline` is approx −0.06
+- `DaysSinceEpoch` is approx −0.0007 (tiny downward trend over time)
+
+Interpretation:
+
+- Topic has a clear effect (especially obama).
+- Sentiment effects are small and slightly negative.
+- The model explains only ~16% of the variance in log‑popularity, confirming that static features alone are weak predictors.
+
+---
+
+### 2.3 Regression Model 2 – Random forest on early time slices
+
+**Goal.** Predict final log‑Facebook popularity for **economy** stories using early Facebook popularity time slices and sentiment.
+
+- **Target:** `log_Facebook` for economy topic, joined with Facebook_Economy time‑series.
+- **Features:**
+
+  - TS1–TS50 (early cumulative popularity counts, cleaned: negative → 0)
+  - `SentimentTitle`, `SentimentHeadline`
+
+We fit a `RandomForestRegressor` with:
+
+- 120 estimators,
+- `min_samples_leaf=2`,
+- `max_depth=None` (trees grow fully),
+- `n_jobs=-1`, `random_state=42`.
+
+**Results (test set):**
+
+- **R² is approx 0.746**
+- **RMSE is approx 0.86** (log‑scale)
+
+Feature importances indicate:
+
+- `TS50` alone contributes ~81% of total importance.
+- Combined sentiment variables contribute ~17%.
+- Earlier TS features each have very small marginal importance.
+
+Thus, knowing an article’s popularity after ~17 hours (TS50) is already highly predictive of its final 2‑day popularity. Early engagement is a much stronger signal than sentiment or publish time.
+
+---
+
+### 2.4 Regression Model 3 – PCA + random forest (dimension reduction)
+
+Model 3 examines the effect of **dimension reduction** on performance.
+
+Instead of using all 50 TS features directly, we:
+
+1. Standardize TS1–TS50 with `StandardScaler`.
+2. Apply PCA with `n_components=10`.
+3. Concatenate the 10 PCA components with the two sentiment features (`SentimentTitle`, `SentimentHeadline`).
+4. Train the same `RandomForestRegressor` as Model 2 on this 12‑dimensional feature space.
+
+PCA results:
+
+- 1st component explains is approx **93.5%** of variance.
+- First 10 components together explain is approx **99.9%** of variance.
+
+**Results (test set):**
+
+- **R² is approx 0.745**
+- **RMSE is approx 0.87**
+
+Compared to Model 2:
+
+- R² decreases only slightly (0.746 → 0.745).
+- RMSE increases minimally (0.862 → 0.865).
+
+So PCA reduces dimensionality from 50 TS features to 10 components with **negligible loss of predictive performance**. The first PCA components effectively summarize overall popularity level and growth pattern, which are the dominant signals for final popularity.
+
+---
+
+### 2.5 Classification Model 4 – Logistic regression for viral vs non‑viral
+
+**Goal.** Classify whether an article is *viral* on Facebook, defined as being in the top 10% of final popularity.
+
+- **Target:**
+
+  - `viral_fb = 1` if `Facebook ≥ 214` (90th percentile), otherwise 0.
+  - Class distribution: ~10% positive, ~90% negative.
+- **Features:**
+
+  - `SentimentTitle`, `SentimentHeadline`
+  - `DaysSinceEpoch`
+  - Topic dummies as before
+
+We intentionally **do not use time‑slice features** here to simulate making a decision at or before publication, when no engagement data is available yet.
+
+We fit a `LogisticRegression` with `max_iter=500` and `class_weight="balanced"` to counter class imbalance.
+
+**Results (test set):**
+
+- **Accuracy is approx 0.73**
+
+  - A naive classifier that always predicts “non‑viral” would obtain is approx 0.90 accuracy, highlighting that raw accuracy is misleading under imbalance.
+- **F1 (viral class) is approx 0.36**
+- **ROC AUC is approx 0.75**
+
+The ROC AUC of 0.75 indicates decent **ranking ability**: the model tends to assign higher probabilities to truly viral articles than to non‑viral ones. However, at the default 0.5 threshold it generates many false positives; tuning the probability threshold would be necessary in practice depending on the business trade‑off between missing viral content and wasting attention on non‑viral items.
+
+---
+
+### 2.6 Clustering Model 5 – K‑means on time‑series shapes
+
+To understand typical growth trajectories of popularity, we cluster early time‑series patterns.
+
+- **Features:** TS1–TS50, standardized with `StandardScaler`.
+- **Sample:** random subset of 5,000 economy+Facebook articles to keep computation manageable.
+- **Algorithm:** `KMeans(n_clusters=3, n_init=10, random_state=42)`.
+
+**Results:**
+
+- **Silhouette score is approx 0.97**, indicating well‑separated clusters (although partly due to one large cluster vs a few small ones).
+- Cluster sizes and mean final Facebook shares:
+
+  | cluster | count | mean shares | median |   max |
+  | ------: | ----: | ----------: | -----: | ----: |
+  |       0 | 4,978 |         ~37 |      3 | 7,045 |
+  |       1 |     1 |       1,886 |  1,886 | 1,886 |
+  |       2 |    21 |      ~2,478 |  1,291 | 8,010 |
+
+Inspecting centroid time‑series (TS1, TS10, TS25, TS50):
+
+- **Cluster 0:** low TS1 (~0.3), slow growth, TS50 is approx 17 → “normal/low popularity” baseline; almost all articles.
+- **Cluster 2:** TS1 is approx 23, TS10 is approx 211, TS50 is approx 1,388 → early rapid take‑off and sustained growth; these are clearly **viral** trajectories.
+- **Cluster 1:** single extreme **super‑viral** outlier with TS1 is approx TS50 is approx 1,886.
+
+Clustering therefore uncovers distinct popularity regimes: ordinary stories, viral stories, and rare super‑viral events.
+
+---
+
+## 3. Decisions and Practical Use
+
+### 3.1 What do the models tell us?
+
+**1. Static metadata is not enough for precise prediction.**
+
+Model 1, using only topic, time and sentiment, explains only about 16% of the variance in log‑Facebook popularity. The EDA already indicated weak correlations between sentiment and engagement, and the model confirms that topic is the only strong static predictor. This means:
+
+- Before any user feedback is observed, we can form only a rough guess about popularity (e.g., “obama stories tend to do better”), but detailed predictions are unreliable.
+
+**2. Early engagement is the key signal.**
+
+Models 2 and 3 show that once ~16 hours of Facebook feedback are available:
+
+- Random forests can explain ~75% of the variance in final log‑popularity.
+- PCA compresses the 50‑dimensional TS inputs to 10 components with essentially no loss in performance.
+
+In practice, this means that **monitoring early time‑series of shares is crucial**. Stories that are already accumulating shares quickly by TS50 are extremely likely to end up as the most popular items after two days.
+
+**3. Logistic regression is useful for ranking, not for definitive labels.**
+
+The viral vs non‑viral classifier has:
+
+- Good ranking ability (ROC AUC ~0.75).
+- Moderate F1 score and relatively low accuracy compared to the majority baseline.
+
+This makes it better suited as a **priority score** than as a hard decision rule. For example, an editorial team might sort draft stories by predicted viral probability to decide where to invest additional editorial resources, but should not automatically discard stories predicted to be non‑viral.
+
+**4. Clustering uncovers growth archetypes.**
+
+K‑means reveals three typical growth shapes:
+
+1. Slow/low growth (most items).
+2. Clearly viral trajectories.
+3. A tiny number of super‑viral events.
+
+Recognizing that an article’s early TS pattern matches the viral or super‑viral cluster can trigger decisions such as:
+
+- Featuring the article more prominently on the homepage.
+- Allocating budget for promoted posts.
+- Producing follow‑up content while interest is high.
+
+### 3.2 How useful are these models for real decisions?
+
+A practical decision workflow informed by this analysis could be:
+
+1. **Pre‑publication / immediately at publication**
+
+   Use the logistic regression model and static features (topic, sentiment, time) to assign each new article a baseline probability of becoming viral. This can help prioritize which stories to monitor more closely, but should not be the sole basis for publication decisions.
+
+2. **Early post‑publication (first few hours)**
+
+   Once some time‑slice information is available (TS1–TS10), use clustering to see whether the article’s early trajectory resembles known viral patterns. Articles already in the viral cluster are good candidates for early promotion.
+
+3. **Mid‑window (around TS50)**
+
+   At ~16–17 hours, feed TS1–TS50 into the PCA + random forest regressor (Model 3) to estimate final reach. This estimate can guide decisions about:
+
+   - How long to keep the story on front pages.
+   - Whether to schedule follow‑ups or derivative content.
+   - Where to allocate marketing/promotional resources.
+
+4. **Limitations**
+
+   - Popularity is still highly stochastic; even with R² is approx 0.75 in the best case, there is considerable residual uncertainty.
+   - Models trained on this dataset focus on four specific topics and a particular time period (2015–2016). Performance may degrade when applied to different domains, languages or time spans. ([arXiv][1])
+
+Overall, these models are best used for **relative ranking and triage** and help in deciding which articles deserve extra attention rather than for exact point predictions of future share counts. Combining static features, early engagement signals, and growth‑pattern clustering yields a practical decision support tool for newsrooms and social media teams working with limited resources.
+
+If you actually read this far...nice! :D
+
+[1]: https://arxiv.org/abs/1801.07055 "Multi-Source Social Feedback of Online News Feeds"

Assignment VI/code.py

@@ -0,0 +1,363 @@
+import zipfile
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import os
+
+from sklearn.model_selection import train_test_split
+from sklearn.linear_model import LinearRegression, LogisticRegression
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.decomposition import PCA
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import (
+	r2_score,
+	root_mean_squared_error,
+	accuracy_score,
+	f1_score,
+	roc_auc_score,
+	confusion_matrix,
+	silhouette_score,
+)
+from sklearn.pipeline import Pipeline
+from sklearn.cluster import KMeans
+
+# ensure imgs dir exists
+os.makedirs("imgs", exist_ok=True)
+
+# data loading
+
+zip_path = "news+popularity+in+multiple+social+media+platforms.zip"
+
+with zipfile.ZipFile(zip_path, "r") as zf:
+	with zf.open("Data/News_Final.csv") as f:
+		news = pd.read_csv(f)
+
+# basic cleaning
+
+pop_cols = ["Facebook", "GooglePlus", "LinkedIn"]
+
+# encode -1 as missing
+for col in pop_cols:
+	news.loc[news[col] < 0, col] = np.nan
+
+# convert publishdate and add numeric time feature
+news["PublishDate"] = pd.to_datetime(news["PublishDate"])
+news["DaysSinceEpoch"] = (
+	news["PublishDate"] - pd.Timestamp("1970-01-01")
+).dt.days
+
+# log transform facebook popularity where available
+news["log_Facebook"] = np.log1p(news["Facebook"])
+
+# eda helpers (optional plotting)
+
+
+def plot_eda():
+	plt.figure()
+	vals = news["Facebook"].dropna()
+	vals = vals[vals > 0]
+	vals.plot.hist(bins=50)
+	plt.xlabel("facebook shares")
+	plt.ylabel("count")
+	plt.title("distribution of facebook popularity")
+	plt.xscale("log")
+	plt.tight_layout()
+	plt.savefig("imgs/eda_facebook_hist.png")
+	plt.close()
+
+	plt.figure()
+	news["log_Facebook"].dropna().plot.hist(bins=50)
+	plt.xlabel("log1p(facebook shares)")
+	plt.ylabel("count")
+	plt.title("distribution of log-transformed facebook popularity")
+	plt.tight_layout()
+	plt.savefig("imgs/eda_log_facebook_hist.png")
+	plt.close()
+
+	mean_by_topic = (
+		news.groupby("Topic")["log_Facebook"].mean().sort_values()
+	)
+	plt.figure()
+	mean_by_topic.plot(kind="bar")
+	plt.ylabel("mean log1p(facebook shares)")
+	plt.title("average facebook popularity by topic")
+	plt.tight_layout()
+	plt.savefig("imgs/eda_mean_by_topic.png")
+	plt.close()
+
+	sample = news.dropna(
+		subset=["log_Facebook", "SentimentTitle"]
+	).sample(5000, random_state=42)
+	plt.figure()
+	plt.scatter(
+		sample["SentimentTitle"],
+		sample["log_Facebook"],
+		alpha=0.3,
+	)
+	plt.xlabel("sentimenttitle")
+	plt.ylabel("log1p(facebook shares)")
+	plt.title("title sentiment vs facebook popularity (sample)")
+	plt.tight_layout()
+	plt.savefig("imgs/eda_sentiment_vs_popularity.png")
+	plt.close()
+
+# model 1: linear regression
+
+
+def run_model_1():
+	df = news.dropna(subset=["log_Facebook"]).copy()
+
+	X = df[["SentimentTitle", "SentimentHeadline", "DaysSinceEpoch", "Topic"]]
+	X = pd.get_dummies(X, columns=["Topic"], drop_first=True)
+	y = df["log_Facebook"]
+
+	X_train, X_test, y_train, y_test = train_test_split(
+		X, y, test_size=0.2, random_state=42
+	)
+
+	linreg = LinearRegression()
+	linreg.fit(X_train, y_train)
+	y_pred = linreg.predict(X_test)
+
+	r2 = r2_score(y_test, y_pred)
+	rmse = root_mean_squared_error(y_test, y_pred)
+
+	print("model 1 – linear regression")
+	print("r2:", r2)
+	print("rmse:", rmse)
+	print("coefficients:")
+	print(pd.Series(linreg.coef_, index=X.columns))
+
+	# optional diagnostic plot
+	plt.figure()
+	plt.scatter(y_test, y_pred, alpha=0.3)
+	plt.xlabel("actual log1p(facebook)")
+	plt.ylabel("predicted log1p(facebook)")
+	plt.title("model 1: actual vs predicted")
+	plt.tight_layout()
+	plt.savefig("imgs/model1_actual_vs_predicted.png")
+	plt.close()
+
+	return linreg, (X_test, y_test, y_pred)
+
+# prepare economy + facebook time-slice data
+
+
+with zipfile.ZipFile(zip_path, "r") as zf:
+	with zf.open("Data/Facebook_Economy.csv") as f:
+		fb_econ = pd.read_csv(f)
+
+# ensure integer id for join
+news["IDLink_int"] = news["IDLink"].astype(int)
+
+news_econ = news[news["Topic"] == "economy"].copy()
+news_econ["IDLink_int"] = news_econ["IDLink"].astype(int)
+
+fb_econ_merged = fb_econ.merge(
+	news_econ, left_on="IDLink", right_on="IDLink_int", how="inner"
+)
+
+# clean time-slice features
+ts_cols = [c for c in fb_econ.columns if c.startswith("TS")]
+for col in ts_cols:
+	fb_econ_merged.loc[fb_econ_merged[col] < 0, col] = 0
+
+# drop rows with missing facebook target
+fb_econ_merged = fb_econ_merged[fb_econ_merged["Facebook"].notna()].copy()
+fb_econ_merged["log_Facebook"] = np.log1p(fb_econ_merged["Facebook"])
+
+ts_cols_early = ts_cols[:50]
+
+# model 2: random forest on raw early ts
+
+
+def run_model_2():
+	X = fb_econ_merged[ts_cols_early + ["SentimentTitle", "SentimentHeadline"]]
+	y = fb_econ_merged["log_Facebook"]
+
+	X_train, X_test, y_train, y_test = train_test_split(
+		X, y, test_size=0.2, random_state=42
+	)
+
+	rf = RandomForestRegressor(
+		n_estimators=120,
+		random_state=42,
+		n_jobs=-1,
+		max_depth=None,
+		min_samples_leaf=2,
+	)
+	rf.fit(X_train, y_train)
+
+	pipe = Pipeline([
+		("scaler", StandardScaler()),
+		("pca", PCA(n_components=10, random_state=42)),
+		("rf", RandomForestRegressor(
+			n_estimators=120,
+			random_state=42,
+			n_jobs=-1,
+			max_depth=None,
+			min_samples_leaf=2,
+		)),
+	])
+
+	pipe.fit(X_train, y_train)
+	y_pred = pipe.predict(X_test)
+
+	r2 = r2_score(y_test, y_pred)
+	rmse = root_mean_squared_error(y_test, y_pred)
+
+	print("model 2 – random forest on raw ts")
+	print("r2:", r2)
+	print("rmse:", rmse)
+
+	importances = pd.Series(rf.feature_importances_, index=X.columns)
+	print("top importances:")
+	print(importances.sort_values(ascending=False).head(10))
+
+	return rf, (X_test, y_test, y_pred)
+
+# model 3: pca + random forest
+
+
+def run_model_3():
+	ts = fb_econ_merged[ts_cols_early]
+	sent = fb_econ_merged[["SentimentTitle", "SentimentHeadline"]]
+	X = pd.concat([ts, sent], axis=1)
+	y = fb_econ_merged["log_Facebook"]
+
+	X_train, X_test, y_train, y_test = train_test_split(
+		X, y, test_size=0.2, random_state=42
+	)
+
+	scaler = StandardScaler()
+	X_train_scaled = scaler.fit_transform(X_train[ts_cols_early])
+	X_test_scaled = scaler.transform(X_test[ts_cols_early])
+
+	pca = PCA(n_components=10, random_state=42)
+	X_train_pca = pca.fit_transform(X_train_scaled)
+	X_test_pca = pca.transform(X_test_scaled)
+
+	train_sent = X_train[["SentimentTitle", "SentimentHeadline"]].values
+	test_sent = X_test[["SentimentTitle", "SentimentHeadline"]].values
+
+	X_train_final = np.hstack([X_train_pca, train_sent])
+	X_test_final = np.hstack([X_test_pca, test_sent])
+
+	rf = RandomForestRegressor(
+		n_estimators=120,
+		random_state=42,
+		n_jobs=-1,
+		max_depth=None,
+		min_samples_leaf=2,
+	)
+	rf.fit(X_train_final, y_train)
+	y_pred = rf.predict(X_test_final)
+
+	r2 = r2_score(y_test, y_pred)
+	rmse = root_mean_squared_error(y_test, y_pred)
+
+	print("model 3 – random forest on pca(ts)")
+	print("r2:", r2)
+	print("rmse:", rmse)
+	print("pca variance explained (first 10):", pca.explained_variance_ratio_)
+	print("total variance explained:", pca.explained_variance_ratio_.sum())
+
+	return rf, (X_test, y_test, y_pred), (pca, scaler)
+
+# model 4: logistic regression (viral vs non-viral)
+
+
+def run_model_4():
+	df = news.copy()
+	df = df[df["Facebook"].notna()].copy()
+
+	threshold = df["Facebook"].quantile(0.9)
+	df["viral_fb"] = (df["Facebook"] >= threshold).astype(int)
+
+	X = df[["SentimentTitle", "SentimentHeadline", "DaysSinceEpoch", "Topic"]]
+	X = pd.get_dummies(X, columns=["Topic"], drop_first=True)
+	y = df["viral_fb"]
+
+	X_train, X_test, y_train, y_test = train_test_split(
+		X,
+		y,
+		test_size=0.2,
+		random_state=42,
+		stratify=y,
+	)
+
+	clf = LogisticRegression(
+		max_iter=500,
+		class_weight="balanced",
+	)
+	clf.fit(X_train, y_train)
+
+	y_pred = clf.predict(X_test)
+	y_proba = clf.predict_proba(X_test)[:, 1]
+
+	acc = accuracy_score(y_test, y_pred)
+	f1 = f1_score(y_test, y_pred)
+	auc = roc_auc_score(y_test, y_proba)
+	cm = confusion_matrix(y_test, y_pred)
+
+	print("model 4 – logistic regression (viral vs non-viral)")
+	print("threshold (shares):", threshold)
+	print("accuracy:", acc)
+	print("f1 (positive class):", f1)
+	print("roc auc:", auc)
+	print("confusion matrix:\n", cm)
+
+	return clf, (X_test, y_test, y_pred, y_proba)
+
+# model 5: k-means clustering on ts shapes
+
+
+def run_model_5():
+	X = fb_econ_merged[ts_cols_early].values
+	scaler = StandardScaler()
+	X_scaled = scaler.fit_transform(X)
+
+	rng = np.random.RandomState(42)
+	idx = rng.choice(X_scaled.shape[0], size=5000, replace=False)
+	X_sample = X_scaled[idx]
+	fb_sample = fb_econ_merged["Facebook"].values[idx]
+
+	kmeans = KMeans(n_clusters=3, random_state=42, n_init=10)
+	kmeans.fit(X_sample)
+	labels = kmeans.labels_
+
+	sil = silhouette_score(X_sample, labels)
+	print("model 5 – kmeans on ts shapes")
+	print("silhouette score:", sil)
+
+	cluster_df = pd.DataFrame(
+		{"cluster": labels, "Facebook": fb_sample}
+	)
+	print(cluster_df.groupby("cluster")["Facebook"].agg(
+		["count", "mean", "median", "max"]
+	))
+
+	centers_scaled = kmeans.cluster_centers_
+	centers = scaler.inverse_transform(centers_scaled)
+	centers_df = pd.DataFrame(centers, columns=ts_cols_early)
+
+	summary = pd.DataFrame({
+		"cluster": list(range(centers_df.shape[0])),
+		"avg_ts": centers_df.mean(axis=1),
+		"ts1": centers_df["TS1"],
+		"ts10": centers_df["TS10"],
+		"ts25": centers_df["TS25"],
+		"ts50": centers_df["TS50"],
+	})
+	print("cluster centroid summary:\n", summary)
+
+	return kmeans, scaler, summary
+
+
+if __name__ == "__main__":
+	run_model_1()
+	run_model_2()
+	run_model_3()
+	run_model_4()
+	run_model_5()
+	plot_eda()

Assignment VI/out.log

@@ -0,0 +1,53 @@
+model 1 – linear regression
+r2: 0.1566089012155698
+rmse: 1.8625218879551908
+coefficients:
+SentimentTitle      -0.383499
+SentimentHeadline   -0.064708
+DaysSinceEpoch      -0.000678
+Topic_microsoft      0.101848
+Topic_obama          1.779152
+Topic_palestine      0.023738
+dtype: float64
+model 2 – random forest on raw ts
+r2: 0.7441325592979975
+rmse: 0.8661035218490399
+top importances:
+TS50                 0.810814
+SentimentHeadline    0.099992
+SentimentTitle       0.067386
+TS49                 0.001883
+TS48                 0.000589
+TS15                 0.000503
+TS18                 0.000503
+TS13                 0.000498
+TS24                 0.000498
+TS10                 0.000480
+dtype: float64
+model 3 – random forest on pca(ts)
+r2: 0.7442278904925559
+rmse: 0.8659421602173341
+pca variance explained (first 10): [9.38529911e-01 3.24317512e-02 1.76049987e-02 7.50439628e-03
+ 1.90148973e-03 6.83679307e-04 3.57135169e-04 2.12058930e-04
+ 1.33577763e-04 9.66846072e-05]
+total variance explained: 0.9994556829781833
+model 4 – logistic regression (viral vs non-viral)
+threshold (shares): 214.0
+accuracy: 0.7287481626653601
+f1 (positive class): 0.35709101466105386
+roc auc: 0.7530964866530827
+confusion matrix:
+ [[10669  4023]
+ [  406  1230]]
+model 5 – kmeans on ts shapes
+silhouette score: 0.9732852082508215
+         count         mean  median     max
+cluster                                    
+0         4978    36.751708     3.0  7045.0
+1            1  1886.000000  1886.0  1886.0
+2           21  2477.761905  1291.0  8010.0
+cluster centroid summary:
+    cluster       avg_ts          ts1         ts10         ts25         ts50
+0        0     8.317766     0.297710     2.959221     7.836079    17.221977
+1        1  1885.920000  1885.000000  1886.000000  1886.000000  1886.000000
+2        2   640.917143    22.761905   211.142857   579.047619  1387.619048

Assignment VI/todo.md

@@ -0,0 +1,25 @@
+Conduct the following analysis for the dataset:
+1. Exploratory Data Analysis
+Explore the statistical aspects of the dataset. Analyze the
+distributions and provide summaries of the relevant statistics. Perform any cleaning,
+transformations, interpolations, smoothing, outlier detection/ removal, etc. required on the
+data. Include figures and descriptions of this exploration and a short description of what
+you concluded (e.g. nature of distribution, indication of suitable model approaches you
+would try, etc.) Min.1 page text + graphics (required).
+
+2. Model Development, Validation and Optimization
+Develop and evaluate three (4000-level) or four (6000-level) or more J models. If possible,
+these models should cover more than one objective, i.e. regression, classification,
+clustering. Consider the efect of dimension reduction of the dataset on model
+performance. Diferent models means diferent combinations of an algorithm and a
+formula (input and output features). The choice of independent and response variables is
+up to you. Explain why you chose them. Construct the models, test/ validate them. Briefly explain the
+validation approach. You can use any method(s) covered in the course. Include your code
+in your submission. Compare model results if applicable. Report the results of the model
+(fits, coeficients, sample trees, other measures of fit/ importance, etc., predictors and
+summary statistics). Min. 2 pages of text + graphics (required).
+
+3. Decisions
+Describe your conclusions from the model
+fits, predictions and how well (or not) it could be used for decisions and why. Min. 1/2 page
+of text + graphics.