[dbjapan] Introducing PAMI (an open-source Python toolkit for pattern mining purposes)

[uday kiran <uday.rage [at] gmail.com>]

Dear Database Society of Japan members,

Good day!

               

                Clustering, Classification, Prediction, and Pattern mining are four pillars of knowledge discovery in databases.  Many open-source Python tools exist to perform clustering, classification, and prediction tasks. Unfortunately, no proper tools exist to perform pattern mining tasks to our knowledge.

             To promote the field of pattern mining, we (myself, Prof. Masashi Toyoda, and Prof. Masaru Kitsuregawa, the University of Tokyo) have developed an open-source python library called PAttern MIning (PAMI).

            PAMI offers more than 70 pattern mining algorithms covering vast pattern types hidden in six different types of databases across multiple computing architectures. In addition, PAMI offers a command line interface, IDE support, Jupyter notebook support, and easy maintenance through Python Package Index.

       Interested researchers are requested to look at the following:

1) URL of the PAMI repository: https://github.com/udayRage/PAMI

2) Tutorials: 

      a) https://towardsdatascience.com/hello-i-am-pami-937439c7984d

 

      b) https://towardsdatascience.com/introduction-to-periodic-frequent-pattern-mining-using-pami-python-library-49a6e9dbc2e5

      c) https://towardsdatascience.com/tensorflow-keras-statsmodel-sklearn-etc-pami-c7210afa62d0

 

         Finally, we invite fellow pattern mining/data mining researchers to collaborate on developing the PAMI library.   

                

--

Regards

Prof. Rage Uday Kiran

Associate Professor

Database Systems Lab

Division of Information Systems

University of Aizu, Aizu-Wakamatsu, Japan. 

Lab number: 145-A

Mobile no. +81 80 4834 0999

Phone no. (Lab) +81 242 37 2541

                  (Ext) 3115

----------------------------------------------------------------------
When people throw stones at you, convert them into milestones.