Center for Machine Learning and Intelligent Systems
About  Citation Policy  Donate a Data Set  Contact

Repository Web            Google
View ALL Data Sets

× Check out the beta version of the new UCI Machine Learning Repository we are currently testing! Contact us if you have any issues, questions, or concerns. Click here to try out the new site.

Mechanical Analysis Data Set
Download: Data Folder, Data Set Description

Abstract: Fault diagnosis problem of electromechanical devices; also PUMPS DATA SET is newer version with domain theory and results

Data Set Characteristics:  


Number of Instances:




Attribute Characteristics:

Categorical, Integer, Real

Number of Attributes:


Date Donated


Associated Tasks:


Missing Values?


Number of Web Hits:



Original Owners of Database:

1. F. Bergadano, A. Giordana, L. Saitta
University of Torino, Italy
Corso Svizzera 185, Torino - tel. (39) 11 7712002
e-mail: bergadan '@' itoinfo.bitnet

2. F. Bracadori, D. De Marchi
Sogesta, Localita' Crocicchio, Urbino, Italy


Enichem (Eni), Ravenna through Sogesta (Eni), Urbino.

Data Set Information:

F. Bergadano supplied this database. Each instance contains many components, each of which has 8 attributes. Different instances in this database have different numbers of components. It was impossible to put one instance on one line. He originally had one instance per file, but this makes it difficult to ftp them (imagine ftp'ing 222 or so files!). I bundled the set of 209 instances into a single data file, prefixing each with the line:

===== Instance number 1: =====

where "n" is a number in [1,221]. However, they are NOT, repeat NOT in sequential order. Twelve (12) of the instances are missing. Bergadano supplied these additional 12 instances (numbers 8,12,32,33,66,69,73,152,167,194,203,208) in a "notused" sub-directory. I bundled these up with the same format in the "notused-instances" file.

A quick scan of their file didn't reveal what the purpose is for these twelve instances.

Attribute Information:

0 - dummy (always 1) - used for numbering - ignore
1 - class - classification (1..6, the same for components of one example)
2 - # - component number (integer)
3 - sup - support in the machine where measure was taken (1..4)
4 - cpm - frequency of the measure (integer)
5 - mis - measure (real)
6 - misr - earlier measure (real)
7 - dir - filter, type of the measure and direction:
8 - omega - rpm of the machine (integer, the same for components of one example)

Relevant Papers:

F. Bergadano, A. Giordana, L. Saitta, F. Brancadori, D. De Marchi: "Integrated Learning in a real Domain" Proc. VII ML Conference, Austin TX, 1990 (pages 322-329)
[Web Link]

Citation Request:

Please refer to the Machine Learning Repository's citation policy

Supported By:

 In Collaboration With:

About  ||  Citation Policy  ||  Donation Policy  ||  Contact  ||  CML