Limits...
Which gene did you mean?

Mons B - BMC Bioinformatics (2005)

Bottom Line: In general people will think about semantic tagging as just another form of text mining, and that term has quite a negative connotation in the minds of some biologists who have been disappointed by classical approaches of text mining.Although remarkable results have been obtained in experimental circumstances, the wide spread use of information mining tools is lagging behind earlier expectations.This commentary proposes to make semantic tagging an integral process to electronic publishing.

View Article: PubMed Central - HTML - PubMed

Affiliation: Biosemantics Group Rotterdam, Department of Medical Informatics, Erasmus MC - University Medical Center Rotterdam, P,O, Box 1738, NL-3000 DR Rotterdam, the Netherlands. b.mons@erasmusmc.nl

ABSTRACT
Computational Biology needs computer-readable information records. Increasingly, meta-analysed and pre-digested information is being used in the follow up of high throughput experiments and other investigations that yield massive data sets. Semantic enrichment of plain text is crucial for computer aided analysis. In general people will think about semantic tagging as just another form of text mining, and that term has quite a negative connotation in the minds of some biologists who have been disappointed by classical approaches of text mining. Efforts so far have tried to develop tools and technologies that retrospectively extract the correct information from text, which is usually full of ambiguities. Although remarkable results have been obtained in experimental circumstances, the wide spread use of information mining tools is lagging behind earlier expectations. This commentary proposes to make semantic tagging an integral process to electronic publishing.

Show MeSH
The part of text of this editorial that was marked in italics was fed to an existing semantic tagging tool and it can be seen in the figure how different expressions are mapped to the same concept number.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC1173089&req=5

Figure 1: The part of text of this editorial that was marked in italics was fed to an existing semantic tagging tool and it can be seen in the figure how different expressions are mapped to the same concept number.

Mentions: In my opinion, author-interactive publication tools should not force the writing scientist to use specific terms from pre-structured lists and nomenclatures. Trying to do exactly that, is what made most efforts of nomenclature standardization ineffective. In contrast, the very valuable efforts of the HGNC's, Entrezgenes and SwissProts of this world should be used to disambiguate terms on the fly and only ask the author for assistance in the rare cases the ontology driven system can not make an informed decision about the meaning of a term, for example in case a homonym is used without sufficiently distinguishing context surrounding it. If a title such as: 'Epidemiological considerations of BSE' is typed, the system is unlikely to be able to decide whether the author meant Breast Self Examination or Bovine Spongiform Encephalopathy. If no further context follows, the author could be prompted to resolve the ambiguity. However, if someone decides that he likes EBV more than Human Herpesvirus 4, or she prefers CD154 over TNFSF5, the author should not be forced to change that in the text, as long as the semantic tag added to that term in the background is linked to the unique concept identifier of the virus or the gene in the leading ontologies and nomenclature data bases (see figure 1). Obviously, some extra work will be asked from the author, but not anywhere near as much as most critics of the semantic web idea seem to expect.


Which gene did you mean?

Mons B - BMC Bioinformatics (2005)

The part of text of this editorial that was marked in italics was fed to an existing semantic tagging tool and it can be seen in the figure how different expressions are mapped to the same concept number.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC1173089&req=5

Figure 1: The part of text of this editorial that was marked in italics was fed to an existing semantic tagging tool and it can be seen in the figure how different expressions are mapped to the same concept number.
Mentions: In my opinion, author-interactive publication tools should not force the writing scientist to use specific terms from pre-structured lists and nomenclatures. Trying to do exactly that, is what made most efforts of nomenclature standardization ineffective. In contrast, the very valuable efforts of the HGNC's, Entrezgenes and SwissProts of this world should be used to disambiguate terms on the fly and only ask the author for assistance in the rare cases the ontology driven system can not make an informed decision about the meaning of a term, for example in case a homonym is used without sufficiently distinguishing context surrounding it. If a title such as: 'Epidemiological considerations of BSE' is typed, the system is unlikely to be able to decide whether the author meant Breast Self Examination or Bovine Spongiform Encephalopathy. If no further context follows, the author could be prompted to resolve the ambiguity. However, if someone decides that he likes EBV more than Human Herpesvirus 4, or she prefers CD154 over TNFSF5, the author should not be forced to change that in the text, as long as the semantic tag added to that term in the background is linked to the unique concept identifier of the virus or the gene in the leading ontologies and nomenclature data bases (see figure 1). Obviously, some extra work will be asked from the author, but not anywhere near as much as most critics of the semantic web idea seem to expect.

Bottom Line: In general people will think about semantic tagging as just another form of text mining, and that term has quite a negative connotation in the minds of some biologists who have been disappointed by classical approaches of text mining.Although remarkable results have been obtained in experimental circumstances, the wide spread use of information mining tools is lagging behind earlier expectations.This commentary proposes to make semantic tagging an integral process to electronic publishing.

View Article: PubMed Central - HTML - PubMed

Affiliation: Biosemantics Group Rotterdam, Department of Medical Informatics, Erasmus MC - University Medical Center Rotterdam, P,O, Box 1738, NL-3000 DR Rotterdam, the Netherlands. b.mons@erasmusmc.nl

ABSTRACT
Computational Biology needs computer-readable information records. Increasingly, meta-analysed and pre-digested information is being used in the follow up of high throughput experiments and other investigations that yield massive data sets. Semantic enrichment of plain text is crucial for computer aided analysis. In general people will think about semantic tagging as just another form of text mining, and that term has quite a negative connotation in the minds of some biologists who have been disappointed by classical approaches of text mining. Efforts so far have tried to develop tools and technologies that retrospectively extract the correct information from text, which is usually full of ambiguities. Although remarkable results have been obtained in experimental circumstances, the wide spread use of information mining tools is lagging behind earlier expectations. This commentary proposes to make semantic tagging an integral process to electronic publishing.

Show MeSH