Extract drug names

t_klok

I am a medical doctor and doing research.

I have an excel sheet with freetext wich contains drugs names.

I want to filter out these drug names and count how many drugs are noted in each field (excel cell).

Any suggestions??

MartinLiebig

Hi@t_klok,

this is one of the problems were i started with "hey that's easy" and it turned out to be a 15operator process. Maybe there is another way to do this?@sgenzermight find one. Anyway, my solution is attached.

You might want to link up with@SvenVanPoucke. He is a physician and our medical expert in the community.

Best,

Martin

<描述一致=“中心”颜色=“透明”有限公司lored="false" width="126">Dummy data for drug texts you can replace this with read excel


<描述一致=“中心”颜色=“透明”有限公司lored="false" width="126">Att needs to be text to work with Process Documents

<运营商激活= " true " class = "特克斯t:process_document_from_data" compatibility="8.1.000" expanded="true" height="82" name="Process Documents from Data" width="90" x="313" y="34">



<运营商激活= " true " class = "特克斯t:tokenize" compatibility="8.1.000" expanded="true" height="68" name="Tokenize" width="90" x="45" y="34"/>

<参数键= "属性" value = "drugname"/>
<描述一致=“中心”颜色=“透明”有限公司lored="false" width="126">only use specifed drug names







<描述一致=“中心”颜色=“透明”有限公司lored="false" width="126">Generate bag of words






























<描述一致=“中心”颜色=“透明”有限公司lored="false" width="126">Dummy data for drug names. You can replace this with read excel





<描述一致=“中心”颜色=“透明”有限公司lored="false" width="126">id will become header in transpose




<描述一致=“中心”颜色=“透明”有限公司lored="false" width="126">Only let attributes through which were present in the lower exa

t_klok

Hi Martin,

Rapid(miner) answers..

Thx I think I understand.

But I would like to filter out drugnames using a list which contains the drugnames.
I do not want to enter all the reference drugnames by hand....

MartinLiebig

Hi,

sure you can just read in the Excel file instead of generating them by hand. That was just to generate some dummy data.

Best,

Martin

sgenzer

hi@t_klok- I'd want to see the data before really weighing in but just from what you describe I would use the Text Processing extension, tokenize, and then Filter Tokens (Dictionary) with the drug names. It's very similar to what@mschmitzbuilt with his XML.

Scott

DocMusher

Hi each country provides a list with official drug names. Additionally, SNOMED can help you find drug names in a text.

---

@sgenzerwrote:

hi@t_klok- I'd want to see the data before really weighing in but just from what you describe I would use the Text Processing extension, tokenize, and then Filter Tokens (Dictionary) with the drug names. It's very similar to what@mschmitzbuilt with his XML.

Scott

---

DocMusher

Hi,

Please take a look at the technology Microsoft is testing:https://www.youtube.com/watch?v=c6exHAzNwy4#action=share

Cheers Sven

t_klok

Thank you all.

I have a (large) list of drugnames and I want to see if freetext fields in an xcl contain any of these names.

So I query an xcl file with freetext cells and the reference is a file with all drugnames.

I do not want to enter all these drugnames one by one in rapidminer.