I have over 2,000 PDFs that I need to extract information from. This requires parsing the PDF and populating known fields. There are several potential formats the form comes in (see attachments) however the text is always the same which preceeds the information of interest. Ideally, the program could extract data from documents which are scanned (ie a scanned fax) however if it only works with embedded text PDFs that is acceptable. Ideally the program will be written in Python, however if there is a compelling reason to write in another language I am open to alternatives.
Please see the three png files (MYR Form 604 example, Third Type and Three Dates Example) for the fields i am trying to extract.
Fields required (as per example document):
Company Name, ACN
1) Substantial Holder name, Substantial holder ACN, Change in interest date, previous notice date, previous notice dated
2) Previous Notice Persons votes, previous notice voting power, present notice persons votes, present notice voting power
3) Date of change, person whose relevant interest changed, nature of change, consideration given in relation to change, class and number of securities affected, persons votes affected
4) Holder of relevant interest, registered holder of securities, person entitled to be registered as holder, nature of relevant interest, class and number of securities, persons votes
5) Changes in association: Name and ACN, Nature of Association
6) Addresses: Name, Address
Many will contain an appendix – I do not need to collect any information from these as they are not standardized.
I have uploaded examples of the pdf files (PDF_Examples), an example of a parser (Parser_Example) and an example of the output (CSV_PDFs) that I am getting now.
Hello there i've previously worked on a project similar to this... i can use python from which you can introduce templates and the respective fields would be extracted