search.txt The searching command we using on Pubmed pids.txt All the pids for fulltext fulltext-crawler.py get fulltext papers we need from pubmed pubmed-crawler.py get abstract,keywords…… from pubmed papers.csv Data we get from above code sorted_words.txt All the key words appear in abstract, we combine them into one-gram (like: Machine Learning -> Machine_Learning) word2vec_batch.py train word2vec model from fulltext, abstract and PMC OA(more than 2 millions papers), when we are training, we combining all the words above fulltext_abstract_phrases3.model word2vec model we got top200cs_combine.txt Top 100 frequency cs key word, with manually cleaning and combining top200medical_combine.txt Top 200 frequency medical key word, with manually cleaning and combining medical1000_combine.txt Top 500 frequency medical key word, with manually cleaning and combining cs1000_combine.txt Top 1000 frequency cs key word, with manually cleaning and combining heatmap_prediction_plot.py Get each pair of real value and predicted value heatmap_clustermap.py Draw the cluster heatmap heatmap_clustermap_2019_pred.py Using all the previous data to predict the number of publications of each pair of CS and medical keywords, and plot the cluster cluster heatmap 2019-6-actal_3.pdf actual heatmap of 2019.1-2019.6 2019-6-pred_3.pdf predicted heatmap of 2019.1-2019.6 heatmap_prediction_all_by_real_sample_from_2010.py Predict every years publications from 2010 to 2019 using all the previous data heatmap_prediction_from_2015.py Show how the R square decrease from 2015 if we using heatmap of 2015 and perdict iteratively heatmap_top20_each_5years.py To get TOP20 most popular CS and medical keywords every 5 years heatmap_top20_in+decrease_each_5years.py To get TOP20 increase and decrease of combination of CS&medical words heatmap_top20_in+decrease_each_5years2.py To get TOP20 increase and decrease of cs words heatmap_top20_in+decrease_each_5years3.py To get TOP20 increase and decrease of medical words heatmap_top20_in+decrease_each_5years4.py To get the number of publications every 5 years in_decrease_for_cs.txt overall top 20 increasing and decreasing only for cs words every 5 years in_decrease_for_med.txt overall top 20 increasing and decreasing only for medical words every 5 years Number_of_publications_each_5_year.txt the number of publications we get by searching keywords of "machine learning" "artificial intelligence" “classifier” “deep learning” "data mining" every 5 years pubmed_figure.ipynb Jupyter notebook for generating all the figures difference between true predict 1. top 20 combinations of AI technologies and biomedical increasing in last 5 years 2. top 20 combinations of AI technologies and biomedical decreasing in last 5 years 3. predicted top 20 combinations increasing in future 5 years 4. predicted top 20 combinations decreasing in future 5 years We need to speculate as to why we think the predicted decrease/increase will happen