{"id":42121,"date":"2023-01-12T18:53:55","date_gmt":"2023-01-12T18:53:55","guid":{"rendered":"https:\/\/www.rightsdirect.com\/?post_type=blog_post&p=42121"},"modified":"2023-02-16T13:49:33","modified_gmt":"2023-02-16T13:49:33","slug":"semantische-suche-vs-schluesselwortsuche","status":"publish","type":"blog_post","link":"https:\/\/www.rightsdirect.com\/de\/blog\/semantische-suche-vs-schluesselwortsuche\/","title":{"rendered":"Semantische Suche vs. Schl\u00fcsselwortsuche"},"content":{"rendered":"\n


Haben Sie schon einmal versucht, mit einer Standardsuchmaschine nach medizinischen Unterlagen zu suchen? Und wenn ja, waren Sie zufrieden mit den erzielten Ergebnissen? Wahrscheinlich nicht. Denn es gibt ernsthafte Einschr\u00e4nkungen bei der Verwendung der Stichwortsuche in der pharmazeutischen Industrie.<\/h2>\n\n\n\n

Stellen Sie sich vor, Sie suchen Papiere mit dem Enzymtyp \u201eGSK\u201c. Wenn Sie eine generische Suchmaschine verwenden, erhalten Sie Artikel, in denen \u201eGlykogensynthasekinase\u201c erw\u00e4hnt wird, sowie Artikel \u00fcber das Unternehmen \u201eGlaxoSmithKline\u201c, das f\u00fcr Ihre Suche hier nicht besonders relevant ist. Eine semantische Suchmaschine, die auf wissenschaftlichen Vokabularen und einem Begriffskl\u00e4rungssystem basiert, konzentriert sich jedoch nur auf Ergebnisse, die das Protein enthalten, und gibt Ihnen Kontextspezifit\u00e4t.<\/h2>\n\n\n\n

Lesen Sie weiter und erfahren Sie, wie semantische Suche Ihnen dabei hilft, die Informationen zu finden, die wirklich wichtig sind.<\/h2>\n\n\n\n


If you needed even more accuracy and wanted to find a specific protein such as GSK3, you would be required to do a search for:<\/p>\n\n\n\n

glycogen synthase kinase 3 alpha, GSK-3-A, GSK3A, alpha glycogen synthase kinase-3, glycogen synthase kinase-3A\u2026<\/p>\n\n\n\n

It\u2019s a pretty long list of synonym derivatives, right?\u00a0 A good semantic search system on the other hand, does all this for you when it indexes so that you don\u2019t have to worry when searching.<\/p>\n\n\n\n

Transformative Data Integration<\/h3>\n\n\n\n

Having done this, you are then set up for better downstream data analysis because your conversion from unstructured to structured (typed) data is way more accurate.<\/p>\n\n\n\n

You can then connect your enriched, structured data to databases and other systems, giving enhanced data connectivity across the organisation and speeding up analysis.<\/p>\n\n\n\n

Group Level Searches<\/h3>\n\n\n\n

Great semantic search provides taxonomic relationships between its entities, so higher-order searches are possible.\u00a0 Let\u2019s take the example of \u2018Viagra\u2019 \u2013 whose current use was found as an adverse effect during its trials for pulmonary hypertension.<\/p>\n\n\n\n

I\u2019d find a bunch of articles that would mention things like Viagra\u2019s protein target, Phosphodiesterase 5A (PDE5A).\u00a0 The image below shows how PDE5A and Phosphodiesterase 11A (PDE11A) were found in an article and where they sit in the taxonomy.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

We can see that PDE5A sits in an enzyme taxonomy under the wider \u2018Phosphodiesterase\u2019 class. I could click on the \u2018Phosphodiesterase\u2019 class and get the system to search for anything under it:<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

You can see how PDE8B and PDE10A were identified in this way.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

This becomes incredibly useful, say if you\u2019re interested in finding out which competitors have developed drugs for a target you\u2019re working on.<\/p>\n\n\n\n

What you\u2019re looking for is a rich set of taxonomies covering areas such as diseases, drugs, protein classes and so on.<\/p>\n\n\n\n

A good semantic search engine will actually embed the concepts (that\u2019s to say, entities such as \u201cPDE5A\u201d, entity classes, e.g. \u201cgene\u201d, or higher level abstractions like \u201cprotein class\u201d) within the plain text.  How is this useful?  Well, query time is really quick and extremely accurate, all because you don\u2019t have to do synonym expansion.<\/p>\n\n\n\n

That\u2019s hard to do in a generic search engine that doesn\u2019t leverage life science taxonomy data in one step.<\/p>\n\n\n\n

Connections<\/h3>\n\n\n\n

Additionally, you could examine the co-occurrence data to get a feel for the landscape.\u00a0 In this example, I could look at the indications commonly associated with documents mentioning PDE5A:<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Here, we quickly see that Erectile Dysfunction and Pulmonary Hypotension are associated with PDE5A \u2013 and also how much time this can save when working in drug repurposing.<\/p>\n\n\n\n

You could also look at co-occurrences on a sentence level.\u00a0 Sentence-level co-occurrences are stronger indicators of a real association between entities than document level.\u00a0 Why? Because at a document level you might find entities in keywords section that hold spurious and unrelated terms.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

A comprehensive autocomplete index helps guide your searches.\u00a0 A little bit more in depth than GSK the company or GSK the protein!<\/p>\n\n\n\n

Fit-For-Science Search<\/h3>\n\n\n\n

But you\u2019re not limited to entities and types that have already been curated.  You can build your own vocabularies or use plain text.<\/p>\n\n\n\n

Remember that semantically enabled search is as good as the vocabularies it\u2019s built on.  An excellent vocabulary with a huge number of synonyms means that typing in the brand name of a drug also brings up papers associated with its clinical name.<\/p>\n\n\n\n

And there you have it \u2013 pitted against the depth and breadth that semantic search offers, keyword search simply cannot compete in terms of accuracy, full awareness, or efficiency.  Semantic search allows you to buy back valuable time that would otherwise be spent sifting through huge amounts of documents, and even convert textual data into something you can integrate across your systems, thanks to entity recognition.<\/p>\n\n\n\n

Ready to learn more? Check out:<\/strong><\/h4>\n\n\n\n