All the above have some common ground. In your opinion, what is the real difference between these fields? if you were asked what do you classify your research area what would that be?
In layman's language, statistics is a way to infer patterns from data based on existing model; machine learning is a heuristics to have the computer form its own model from the data; data mining and pattern recognition are applications (not methods) that can be done through either statistics or machine learning; and pattern recognition is a sub-field of data mining. Many people would just claim they do all of them, I guess.
I do woodworking and carpentry using routers and saws, etc., BTW ;)
In layman's language, statistics is a way to infer patterns from data based on existing model; machine learning is a heuristics to have the computer form its own model from the data; data mining and pattern recognition are applications (not methods) that can be done through either statistics or machine learning; and pattern recognition is a sub-field of data mining. Many people would just claim they do all of them, I guess.
I do woodworking and carpentry using routers and saws, etc., BTW ;)
For me, data mining is a process that discover useful and surprising knowledge from data. Data miners get raw data from users and users may ask them questions:
Tell me what is important in the data? in this case we have frequent pattern or association rule mining.
Tell what is unexpected or surprising in the data? in this case we have outlier, change or abnormal detection.
I want to see something about the data? in this case we have visual analytics.
Many useful knowlegdes discovered from the data are then exploited for building prediction, recommendation or classification models.
I think that the difference is basically an historical one.
Statistics is the earliest of these 4 fields, first coming as applied Mathematics. There are works on classification in the beginning of the 19th century (even before Fisher's 1936 seminal paper on "the use of multiple measurements in taxonomic problems").
Then came Pattern Recognition (PR), in a period (the 1970's) where Computer Science was centered on perception problems (OCR, Speech Recognition, image Processing,...). Machine Learning (ML) appeared in the 1980's as an Artificial Intelligence field.
Data Mining (DM) appeared later, as a subfield of Data Base Engineering.
Of course, from the functional point of view, Ji He is right as PR and DM can be considered as applications of ML, as well as ML can be considered as application of Statistics to Computer Science.
- Badges
- Science topic