diff --git a/Project/simple_approach/Evaluation_sentiment_dataset.ipynb b/Project/simple_approach/Evaluation_sentiment_dataset.ipynb
new file mode 100644
index 0000000..8c3f7f2
--- /dev/null
+++ b/Project/simple_approach/Evaluation_sentiment_dataset.ipynb
@@ -0,0 +1,321 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "%matplotlib inline"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Using TensorFlow backend.\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[nltk_data] Downloading package punkt to /home/jonas/nltk_data...\n",
+ "[nltk_data] Package punkt is already up-to-date!\n",
+ "[nltk_data] Downloading package averaged_perceptron_tagger to\n",
+ "[nltk_data] /home/jonas/nltk_data...\n",
+ "[nltk_data] Package averaged_perceptron_tagger is already up-to-\n",
+ "[nltk_data] date!\n",
+ "[nltk_data] Downloading package wordnet to /home/jonas/nltk_data...\n",
+ "[nltk_data] Package wordnet is already up-to-date!\n"
+ ]
+ }
+ ],
+ "source": [
+ "import numpy as np \n",
+ "import pandas as pd \n",
+ "import simple_twitter_learning as stl\n",
+ "import re"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "* download data"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "dataset already downloaded\n"
+ ]
+ }
+ ],
+ "source": [
+ "%%bash\n",
+ "\n",
+ "if [ ! -e 'dataset_sentiment.csv' ]\n",
+ "then\n",
+ " echo \"downloading dataset\"\n",
+ " wget https://raw.githubusercontent.com/SmartDataAnalytics/MA-INF-4222-NLP-Lab/master/2018_SoSe/exercises/dataset_sentiment.csv\n",
+ "else\n",
+ " echo \"dataset already downloaded\"\n",
+ "fi"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "df = pd.read_csv('dataset_sentiment.csv')\n",
+ "df = df[['text','sentiment']]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "
\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " text | \n",
+ " sentiment | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 0 | \n",
+ " nancyleegrahn how did everyone feel about th... | \n",
+ " Neutral | \n",
+ "
\n",
+ " \n",
+ " 1 | \n",
+ " scottwalker didnt catch the full gopdebate l... | \n",
+ " Positive | \n",
+ "
\n",
+ " \n",
+ " 2 | \n",
+ " tjmshow no mention of tamir rice and the gop... | \n",
+ " Neutral | \n",
+ "
\n",
+ " \n",
+ " 3 | \n",
+ " robgeorge that carly fiorina is trending ho... | \n",
+ " Positive | \n",
+ "
\n",
+ " \n",
+ " 4 | \n",
+ " danscavino gopdebate w realdonaldtrump deliv... | \n",
+ " Positive | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
"
+ ],
+ "text/plain": [
+ " text sentiment\n",
+ "0 nancyleegrahn how did everyone feel about th... Neutral\n",
+ "1 scottwalker didnt catch the full gopdebate l... Positive\n",
+ "2 tjmshow no mention of tamir rice and the gop... Neutral\n",
+ "3 robgeorge that carly fiorina is trending ho... Positive\n",
+ "4 danscavino gopdebate w realdonaldtrump deliv... Positive"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "df['text'] = df['text'].apply(lambda x: x.lower())\n",
+ "df['text'] = df['text'].apply(lambda x: x.replace('rt',' '))\n",
+ "df['text'] = df['text'].apply((lambda x: re.sub('[^a-zA-Z0-9\\s]','',x)))\n",
+ "df.head()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "neg = np.array([df['sentiment'][i] == 'Negative' for i in range(df.shape[0])])\n",
+ "pos = np.array([df['sentiment'][i] == 'Positive' for i in range(df.shape[0])])\n",
+ "neu = np.array([df['sentiment'][i] == 'Neutral' for i in range(df.shape[0])])\n",
+ "\n",
+ "text = np.array(df['text'].tolist())"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "* load pipeline"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pipeline_file = \"/home/jonas/Dokumente/NLP_DATA/python_dumps/pipelines/tfidf_final/final_epoch01.pipeline\"\n",
+ "pm = stl.pipeline_manager.load_from_pipeline_file(pipeline_file)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "* plot statements"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "top_20 = list(\"😳😋😀😌😏😔😒😎😢😅😁😉🙌🙏😘😊😩😍😭😂\")\n",
+ "top_20_sents = stl.emoji2sent(top_20)\n",
+ "\n",
+ "pred_pos = pm.predict(text[pos])\n",
+ "pred_neg = pm.predict(text[neg])\n",
+ "pred_neu = pm.predict(text[neu])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "