General information about the CLUS-RM library

This is the repository with the source code of the CLUS-RM library.

The library contains the code of the CLUS-RM redescription mining algorithm [1,2], based on Predictive Clustering Trees [5]. It also contains the following algorithm extensions: a) using random forest of Predictive Clustering Trees [2] to produce redescriptions, b) using conjunctive refinement procedure [3] to increase redescription accuracy, c) using constraint-based redescrition mining [4] which allows adding user-defined constraints on attributes that build redescription queries. The constraint-based setting allows three modes of targeted redescription exploration.

It also incorporates two modes of redescription set optimization: a) optimization by redescription exchange [1,2] and b) optimization by redescription extraction [3].

This library was created as a result of scientific work of Matej Mihelčić (matmih1@gmail.com) in the Data Mining field called Redescription mining [6] with the goal of obtaining a PhD degree in Computer Science at Jožef Stefan International Postgraduate School, Ljubljana, Slovenia (https://www.mps.si/splet/index.asp?lang=eng).

Implementation

Matej Mihelčić, Ruđer Bošković Institute, Zagreb, Croatia, matmih1@gmail.com

Scientific contributors

Student:

• Matej Mihelčić, Ruđer Bošković Institute, Zagreb, Croatia

Supervisors:

• Dr. Tomislav Šmuc, Ruđer Bošković Institute, Zagreb, Croatia
• Prof. Dr. Nada Lavrač, Jožef Stefan Institute, Ljubljana, Slovenia

Main scientific collaborators:

• Prof. Dr. Sašo Džeroski, Jožef Stefan Institute, Ljubljana, Slovenia

Original publications:

1. M. Mihelčić, S. Džeroski, N. Lavrač, and T. Šmuc, “Redescription mining with multi-target predictive clustering trees,” in Proceedings of the 4th International Workshop, New Frontiers in Mining Complex Patterns, NFMCP 2015, Held in conjunction with ECMLPKDD 2015, Porto, Portugal, September 7, 2015, Revised Selected Papers, M. Ceci, C. Loglisci, G. Manco, E. Masciari, and Z. W. Ras, Eds. Cham: Springer International Publishing, 2016, pp. 125–143, isbn: 978-3-319-39315-5. doi: 10.1007/978-3-319- 39315-5_9. [Online]. Available: http://dx.doi.org/10.1007/978-3-319-39315-5_9.

2. M. Mihelčić, S. Džeroski, N. Lavrač, and T. Šmuc, “Redescription mining augmented with random forest of multi-target predictive clustering trees,” Journal of Intelligent Information Systems, pp. 1–34, 2017, In press, issn: 1573-7675. doi: 10.1007/s10844-017-0448-5. [Online]. Available: http://dx.doi.org/10.1007/s10844-017-0448-5.

3. M. Mihelčić, S. Džeroski, N. Lavrač, and T. Šmuc, “A framework for redescription set construction,” Expert Systems with Applications, vol. 68, pp. 196–215, 2017, issn: 0957-4174. doi: http://doi.org/10.1016/j.eswa.2016.10.012. [Online]. Available: http://www.sciencedirect.com/science/article/pii/S0957417416305437.

4. M. Mihelčić, G. Šimić, M. Babić Leko, N. Lavrač, S. Džeroski, T. Šmuc, and for the Alzheimer’s Disease Neuroimaging Initiative, “Using redescription mining to relate clinical and biological characteristics of cognitively impaired and alzheimer’s disease patients,” PLOS ONE, vol. 12, no. 10, pp. 1–35, 2017, doi: 10.1371/journal.pone.0187364. [Online]. Available: https://doi.org/10.1371/journal.pone.0187364.

---

Related publications:

5. H. Blockeel, “Top-down induction of first order logical decision trees,” PhD thesis, Katholieke Universiteit Leuven, Belgium, 1998.

6. N. Ramakrishnan, D. Kumar, B. Mishra, M. Potts, and R. F. Helm, “Turning CARTwheels: An alternating algorithm for mining redescriptions,” in Proceedings of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Seattle, Washington, USA, 2004, pp. 266–275. doi: 10.1145/1014052.1014083. [Online]. Available: http://doi.acm.org/10.1145/1014052.1014083.