Comparing research trends through author-provided keywords with machine extracted terms: A ML algorithm approach using publications data on neurological disorders

Objective. This study aimed to identify the primary research areas, countries, and organizational involvement in publications on neurological disorders through an analysis of human-assigned keywords. These results were then compared with unsupervised machine-algorithm-based extracted terms from title abstract gain knowledge about deficiencies both techniques. has enabled us understand how far machine-derived titles abstracts can be a substitute for keywords scientific articles. Design/Methodology/Approach. While significant areas identified author-provided downloaded Web Science PubMed, these by based models like VOSviewer techniques YAKE CounterVectorizer. Results/Discussion. We observed that post-covid-19 era witnessed more various disorders, but authors still chose generic keyword list than specific ones. The extraction tool, VOSviewer, many other extraneous insignificant along However, our self-developed machine learning algorithm using CountVectorizer provided precise subject adding stop-words dictionary stop-word NLTK tool kit. Conclusion. although author play vital role as they are assigned broader sense increase readability, concept lacked specificity in-depth analysis. suggested ML being compatible unstructured data was valid alternative author-generated accurate results. Originality/Value. To knowledge, this is first-ever machine-extracted real datasets, which may essential lead domain. Replicating large datasets different fields valuable resource experts stakeholders.