Information extraction, indexing and search of PDF, word and text documents with MongoDB -


does mongodb have feature store pdf, text or .doc/docx documents , search them or match between 2 documents on keyword found in content?

for example:

i might want store 1 document called 'claim.txt' has values for
diagnosis code, short description, date , amount in it.
need store 1 called 'physician_diagnosis.pdf' has, among other text, matching short description in it.

i issue query find document has both matching date , same diagnosis. (e.g. 'pneumonia', '12/12/2012')

is possible mongodb using api, or need pre-processing?

if possible, please point me example , documentation.

your task better suited solr (http://lucene.apache.org/solr/), has inputs many different documents (http://wiki.apache.org/solr/extractingrequesthandler). have write code proper extraction though.

mongodb more meant structured data - although call them documents, not mean "pdf documents" or "word documents" here. it's generic format supports nested field types call document, opposed relational database row doesn't allow that.


Comments

Post a Comment

Popular posts from this blog

SPSS keyboard combination alters encoding -

my question not have programming practical issue on spss. accident keyword combination cannot remember because did not intend press changed encoding in spss , text in data in different language english shown small boxes. nevertheless, when kind of analysis performed different language text shown in output. does know reason? tried change encoding did not produce result. first, @ edit > options , see whether setting unicode or locale. might fix it. if have text in multiple languages, want unicode. otherwise, may have specific locale set wrong. can change set locale, need know locale want use.
Read more

Socket.connect doesn't throw exception in Android -

if put nonsense url no exception thrown , none of rest of code executed not rest of asynctask called method connects. try { socketcliente.connect(new inetsocketaddress("f", port), 2000); } catch (unknownhostexception e) { geterror("host doesn't exist"); return -1; } catch (ioexception e) { geterror("could not connect: host down"); return -1; } add catch statement. catch ( exception e ) { log.d(tag, e.getmessage(),e); toast.maketext(getapplicationcontext(), "unexpected error:" + e.getmessage(), toast.length_long).show(); } this write log , pop toast tell what's going on.
Read more

iphone - How do I keep MDScrollView from truncating my row headers and making my cells look bad? -

Image
here's screenshot show describing: and relevant code: - (nsinteger)spreadview:(mdspreadview *)aspreadview numberofcolumnsinsection:(nsinteger)section{ return self.viewmodel.matrix.width - 1; } - (nsinteger)spreadview:(mdspreadview *)aspreadview numberofrowsinsection:(nsinteger)section{ return self.viewmodel.matrix.height - 1; } -(mdspreadviewcell *)spreadview:(mdspreadview *)aspreadview cellforrowatindexpath:(mdindexpath *)rowpath forcolumnatindexpath:(mdindexpath *)columnpath{ mdspreadviewcell *cell = [aspreadview dequeuereusablecellwithidentifier:@"infocell"]; if (cell == nil){ cell = [[[mdspreadviewcell alloc] initwithstyle:mdspreadviewcellstyledefault reuseidentifier:@"infocell"] autorelease]; } cell.textlabel.text = ((dtvendor *)self.vendors[rowpath.row]).name; return cell; } -(cgfloat)spreadview:(mdspreadview *)aspreadview heightforrowheaderinsection:(nsinteger)rowsection{ return 200; } -(cgfloat)spreadview:(mdspreadview...
Read more