add dataset info

notebooks/vector_space_models.ipynb

@@ -20,6 +20,19 @@
     "from htwgnlp.embeddings import WordEmbeddings"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The embeddings we use for this lab are from the [Google News Word2Vec model](https://code.google.com/archive/p/word2vec/). This model was trained on part of the Google News dataset (about 100 billion words). \n",
+    "\n",
+    "The model contains 300-dimensional vectors for 3 million words and phrases and is about 3.5GB large.\n",
+    "\n",
+    "For this notebook, we use a small subset of 243 words, which were selected beforehand and are stored in the pickle file `data/embeddings.pkl`.\n",
+    "\n",
+    "Besides some sample words, it contains mostly capitals and countries. We will use the embeddings to find analogies between words."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 2,