Update README.md

README.md

@@ -5,7 +5,7 @@ This notebook was used with a collection for the case study on emigration and sh
 
 For classification, topic modelling (LDA) was chosen because it showed the best performance in classification (after experiments with word embeddings or LDA and word embeddings combined). LDA provides a way to group documents by topic and perform similarity searches and improve precision. Thanks to sklearn, it is relatively easy to test different classifiers for a given topic classification task. Logistic regression was chosen as binary classifier. 
 
-*Output graph using an unseen collection on the topic of emigration  (~1000 clippings) .* 
+*Output graph using an unseen collection on the topic of emigration  (790 newspaper clippings).* 
 
 ![Collection on the topic of Emigration](images/categories.PNG)