![Why do my word documents open as read only](https://kumkoniak.com/71.jpg)
![why do my word documents open as read only why do my word documents open as read only](https://www.howtogeek.com/wp-content/uploads/2015/06/04_clicking_browse1.png)
![why do my word documents open as read only why do my word documents open as read only](https://newfare900.weebly.com/uploads/1/2/6/2/126204215/955853455.png)
I don’t know which stemming method professor used (maybe by nltk), it doesn’t seem to be very effective, maybe I can try the tokenizer method in BERT.
- **2.**Use stoplist and stemming both have a positive effect on the improvement of results, and the improvement of stoplist is greater(except in “tfidf”).
- **1.**The best results appear when using stoplist, stemming and tfidf.
- I drawed the following conclusions from the above results.
In order to do the ablation study, I used the controlled variable method. The following are the results of my experiment. Then I calculate the cosine similarity between the doc vector and the query vector, and sort according to the cosine similarity to get the final recall document. "idf" represents the ratio of the total number of documents in the corpus to the total number of documents containing the word.tfidf is the product of "tf" and "idf" It indicates the ratio of the number of times a word appears in the document to the total number of words in the document. According to the term weighting schemes I construct a vector, or embedding, for doc and query. This IR system is very simple, the main method is the vector space model.