changes

README.md

@@ -18,9 +18,11 @@
  ## Instalacja
 1. Sklonuj repozytorium:
    ```bash
-   git clone https://github.com/yourusername/bitsent.git
-   cd bitsent
+   git clone https://github.com/s481877/BitSent.git```
+   cd BitSent/src
 2. Upewnij się że python jest zainstalowany
-    ```python --version```
+    ```bash 
+   python --version
 3. Uruchom skrypt z lokalizacji pliku
-    ```python main.py```
+    ```bash
+    python main.py

src/analysis.py

@@ -1,7 +1,7 @@
 import pandas as pd
 from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
 import requests
-
+import re
 
 def fetch_bitcoin_price():
     url = "https://api.coingecko.com/api/v3/simple/price"
@@ -24,6 +24,13 @@ def load_csv(file_path):
     print(f"CSV loaded. Columns: {df.columns}")
     return df
 
+def clean_text(text):
+    text = re.sub(r'http\S+', '', text)
+    text = re.sub(r'@\w+', '', text)
+    text = re.sub(r'#\w+', '', text)
+    text = re.sub(r'\s+', ' ', text).strip()
+    return text
+
 def get_sentiment(text):
     analyzer = SentimentIntensityAnalyzer()
     sentiment = analyzer.polarity_scores(text)
@@ -53,6 +60,7 @@ def date_format(df):
     df['timestamp'] = pd.to_datetime(df['timestamp'], format='%Y-%d-%m %H:%M:%S')
     df['timestamp'] = df['timestamp'].dt.ceil('2h')
     df['timestamp'] = df['timestamp'].dt.strftime('%Y-%m-%d %H:%M:%S')
+    df['text'] = df['text'].apply(clean_text)
     return df
 
 def btc_data_(date):

src/datasets.ipynb

@@ -30,6 +30,12 @@
    ],
    "execution_count": 49
   },
+  {
+   "metadata": {},
+   "cell_type": "markdown",
+   "source": "In this notebook datasets are processed and merged",
+   "id": "8442de414f013864"
+  },
   {
    "metadata": {
     "ExecuteTime": {