search

kermitt2 / grobid

A machine learning software for extracting information from scholarly documents
https://grobid.readthedocs.io
Apache License 2.0
3.43k stars 443 forks source link

Title is missing ; Results not matching with Web Service and Web Application #332

Open rajeshkumargp opened 6 years ago

rajeshkumargp commented 6 years ago

Hi, I tried to convert PDF to Full Text Document.
The title of the document is missing in the extracted xml document. Here are my trails.

Trail 1:
In Webapp , the title statement is missing. 

<titleStmt>
                <title level="a" type="main"></title>
    </titleStmt>

In Webapp, with Consolidate header option enabled, 

<titleStmt>
                <title level="a" type="main">  Document Title    </title>
    </titleStmt>

Trail 2: From CURL Command,

  1. curl -v --form input=@./ASample.pdf http://172.16.28.52:8900/api/processFulltextDocument

  2. curl -v --form input=@./ASample.pdf consolidateHeader=1 http://localhost:8900/api/processFulltextDocument

  3. curl -v --form input=@./ASample.pdf --form consolidateHeader=1 http://localhost:8900/api/processFulltextDocument

For all three, I got 

 <titleStmt>
                <title level="a" type="main"></title>
    </titleStmt>

Title is missing.

Please guide/suggest me steps to improve/get title fileld in results. Batch mode is also fine.

kermitt2 commented 6 years ago

Hello @rajeshkumargp !

Thanks for reporting the problem, could you add the PDF (or send it to me by email if it is not public) so that we can reproduce the error?

rajeshkumargp commented 6 years ago

Please refer the PDF in the below link. http://www.jpma.org.pk/PdfDownload/8618.pdf

kermitt2 commented 5 years ago

Reading order issue, the title comes at the end of the page in the PDF stream and for some obscure reasons it vanishes in the limbos.