ICFHR 2016 Competition on Recognition of On-line Handwritten Mathematical Expressions (ICFHR-CROHME-2016)
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Symbol Recognition: Classiﬁcation of isolated math symbols
Lot of expression recognition system use an isolated symbol classifier in their complete process. This task compares them. Furthermore, this simple task is a good simple pattern recognition task to compare machine learning solutions.
This task is split into two sub-tasks :
- in Task-2a only valid symbols are provided as input
- in Task-2b incorrect symbol segmentations (junk) are mixed with valid symbols.
For these tasks, each inkML test ﬁle contains a single symbol with an associated identiﬁer (‘UI’ tag). Participants submitted a CSV ﬁle containing one line for each test ﬁle. Each line provides the symbol identiﬁer followed by a ranked list of the Top-10 classiﬁcation candidates.
A CROHMELib tool (evalSymbolIsole.py) was used to compute symbol recognition rates along with the average rank of the correct symbol class (TMP: True Mean Position), where any target class not appearing in the Top-10 is treated as rank 11.
For Task 2b where some inputs were ‘junk’ symbols (i.e.incorrectly segmented symbols), the tool also provides the symbol true positive rate (TAR: True Acceptance Rate) and false positive rate (FAR: False Acceptance Rate).
No comments on this dataset yet.