Research Database Search

[Atul Mishra 14]

Hi, I, as a researcher, have been using the Evernote legacy as a storage and retrieval of information from text books, reference books, etc. of various topics/domains. I have over 1000 of these files in EN. Searching a keyword gives, sometimes, hundreds of files, hence pinpointing the note where required information is present, becomes quite challenging. As of now, what I have done is made a separate file of "Table of contents" of very few important books (by splitting the pages of PDF of those books), and when I have to search, I first search it here in the "Table of Content" of these important books, and if I get lucky, I am able to search the required topic with little ease. Splitting the "Table of Content" of these many books would again be very challenging. Hence, I also tag, as much as possible, the note containing the book with topics given in the book and keep it in respective notebooks (domains). Again, this task is also time-consuming. 

This has been hampering my productivity to a great extent.

If you have any idea to overcome this problem in EN, kindly inform me.

Thanks in advance.

[Mike P 2,981]

The fact that  EN searches within a pdf file is both a blessing and a curse! I would love the ability to exclude attachments from a search which I think would make things easier in your case.

I think the idea of separating out the TOC is an excellent one. Personally I would put each TOC in a separate note (with a link to the note containing the pdf) and either put it in a notebook containing only tables of contents or give it a TOC tag. You can then easily restrict your search to only TOCs which will avoid all those hits where the word is mentioned in the pdf but it's not really a major theme.

[PinkElephant 8,829]

Not exactly what you describe, but 

-contains:filePDF mysearchterm

will exclude all notes from the search result that as well contain a pdf file. It does however not play a role if the searched term is in the pdf, the note text or both.

[eric99 1,083]

On 9/22/2022 at 9:12 AM, Mike P said:

The fact that  EN searches within a pdf file is both a blessing and a curse! I would love the ability to exclude attachments from a search which I think would make things easier in your case.

Good point! The only possibility to exclude pdf is searching in the title only ( intitle:  )

Or use an additional free account which shares all your notes where you may search without pdf 😉

[Mike P 2,981]

1 hour ago, eric99 said:

Or use an additional free account which shares all your notes where you will search without pdf 😉

There really should not be examples, however trivial, where the free plan has capabilities which the paid for plans do not have!

 

3 hours ago, PinkElephant said:

Not exactly what you describe, but 

-contains:filePDF mysearchterm

will exclude all notes from the search result that as well contain a pdf file. It does however not play a role if the searched term is in the pdf, the note text or both.

This is useful provided you do not have text in the note as well as the attached pdf. For me, sometimes that works well, but other times I want a pdf and text. I think currently a good principle is that if you have a very large pdf it is worth considering putting this in a note by iteself and then having a second note with any supporting info - e.g. TOC, notes etc. Obviously each note would need to contain a link to the other note.

[PinkElephant 8,829]

Not being able to search in pdfs is not a feature - even if it helps you in a very specific use case.

One addition (on which I have a support ticket running): On mobile clients while in offline mode, pdf content in offline notebooks is not searched neither.

[Dave-in-Decatur 4,006]

I suspect that Evernote's search of PDFs may not have been designed with a view to searching multiple PDFs each of which contained entire an book. Perhaps, as I believe @Atul Mishra said, splitting the books and putting each chapter in a separate note might be more useful, and either carefully tagging each note, or including specific keywords in the body of each note might also help.

If I may suggest an alternative, unless you need to do this on a mobile phone, there is a research suite called Nota Bene which includes a multilingual word processor designed with scholars in mind, and has a module that does perform very exact searches on PDFs and other documents, and display the context of each found result so you can get to what you're really looking for. It is available for both Windows and Mac.

[gazumped 12,072]

Hi.  I totally agree that the more content you have,  the more likelyhood of a false positive hit on a word in the database.

I tend to try to cure (or at least reduce) that by using phrases and composite searches - a phrase and a single word or specific keyword that is most likely to appear in my target sources.

[Atul Mishra 14]

23 hours ago, Mike P said:

I think the idea of separating out the TOC is an excellent one. Personally I would put each TOC in a separate note (with a link to the note containing the pdf) and either put it in a notebook containing only tables of contents or give it a TOC tag. You can then easily restrict your search to only TOCs which will avoid all those hits where the word is mentioned in the pdf but it's not really a major theme.

Thanks, This is exactly what I have been doing.

It has a certain limitation that if a keyword or phrase is present multiple times inside an important note (for which ToC note has not been made), it brings us to the same hard situation, as agreed by you.

23 hours ago, Mike P said:

The fact that  EN searches within a pdf file is both a blessing and a curse!

Perfect line.

 

 

[Atul Mishra 14]

22 hours ago, PinkElephant said:

It does however not play a role if the searched term is in the pdf, the note text or both

Almost all the notes are having PDF only. 

[Atul Mishra 14]

14 hours ago, Dave-in-Decatur said:

I believe @Atul Mishra said, splitting the books and putting each chapter in a separate note might be more useful, and either carefully tagging each note, or including specific keywords in the body of each note might also help.

Certainly, but as the number of books/notes are over 1000, it would be extremely tough for me to split each book in its chapters. Can we do it automatically using a software? Any Idea.

If not, can we download the ToC from the bookmark section of the PDF viewer automatically in a text file or PDF file. This would be of some help to me in generating the TOC of as many PDF as possible.

14 hours ago, Dave-in-Decatur said:

If I may suggest an alternative, unless you need to do this on a mobile phone, there is a research suite called Nota Bene which includes a multilingual word processor designed with scholars in mind, and has a module that does perform very exact searches on PDFs and other documents, and display the context of each found result so you can get to what you're really looking for. It is available for both Windows and Mac.

Certainly will look into this software. 

Yes I do careful tagging of each note, as much as possible. Thanks.

[Mike P 2,981]

1 hour ago, Atul Mishra said:

can we download the ToC from the bookmark section of the PDF viewer automatically in a text file or PDF file.

I think the copying of the ToC into a note is always going to be a manual operation. Normally a simple "copy and paste" works well - potentially made a bit quicker using the global "paste to Evernote" shortcut (alt+ctrl+V on windows). For text books, the publisher's website for the book will often contain the ToC and you could webclip it from there. For a new book the amount of work you would need to do is not huge, but obviously it will take you a long time to work back through all your over 1000 books.

One other thought, although I'm not sure it will help. You can link to files from an EN note to a file on your hard drive, Google docs or Dropbox. If the pdfs were not in EN then they wouldn't be searched and you could just include the link and the the meta data (tags and ToC) in your EN note. I don't really think that offers any advantages to putting  the pdf into a note by itself and excluding it from the search by notebook, tag or @PinkElephantsearch string suggestion. Just included it here for completeness.

 

[PinkElephant 8,829]

I have several large pdfs in my notes, and I have never seen a problem with book sized files. When searching the notes are found, and the in file search finds the keyword pretty fast, and shows them highlighted for navigation.

There is a problem if several files are in one single note, since search will find the note, but not show in which pdf the keyword is located. So better avoid multiple pdfs in a single note.

If you split a single pdf in several ones (like chapters), I would then put each one into a new note. The TOC will show the notes title, but behind it will be the GUID, the unique identifier each note carries. So just copying the TOC from the book won’t do if you want to use it to navigate in EN. You need to name every note for the chapter title (plus probably some more), and then create the TOC entries from the notes by copying the links. The note titles are then the TOC entries.

I would avoid this sort of work for books or similar documents existing as pdf files - it is plain organizational and does not add true information. Either explore the options for searching discussed above, or try more specific software for research purposes.

[gazumped 12,072]

Just a random thought,  but would an academic literature management app like https://docear.org/ help?