To blacklist a word in Qiqqa (http://www.qiqqa.com/70557), have you put "και" on a line by itself, obviously not in quotes, with nothing else on that line? If so, give the developers your stopwords for a Greek Qiqqa.

@Julie, i think the problem may be the number of autotags in the blacklist.. as it happens the free version only applies 5 tags from the blacklist and whitelist each.. to utilize the full potential of this, the yellow star at the bottom left of the 'manage autotags' pop-up suggests to consider the premium version.. :)

@Julie: Give the developers a list of your Greek stopwords to be inserted and ask them to put it in a free Greek Qiqqa. You cannot, even in a premium version, list all common, uninteresting Greek phrases. It has to go into the software.

Hi Ian,

I am getting a little confused with the way expedition generates theme for me.. when it says a certain number of themes have been generated, there are quite frequently duplicate themes spelled exactly the same but containing different articles... i'm not sure if the difference is whether a document is a digital pdf vs an ocr reading of an older pdf file.. is there a way i can merge two themes created in expedition.

regards, raza

@Raza: Yes, clustering produces some "interesting" results. My example is the theme "Data". Its not related to OCR'ing. You may find duplicated names for themes, but with different tags inside. It is less likely to happen if you ask for a smaller number of themes.

It might be that there are two groups of researchers (on different continents or from different schools of thought), using different vocabulary. Inspect some articles with the search capability. If you want to combine duplicate-looking themes, you can move them to a separate library, especially worthwhile if it is a bigger theme than average.

@Ian: Thank you so much... it is a very interesting point and actually can bring a lot of insight into the reasons for articles on the same topic being grouped into different themes.. I hadn't thought on these lines but it would be very nice to have my articles segregated on some academic divider, especially different schools of thought... thanks for pointing this out..

regards, raza

Hi Ian, would appreciate your help in another problem i'm experiencing...

I am starting a new study using qiqqa and searching for literature on the qiqqa web browser... I have retrieved quite a few relevant articles on pubmed using my search terms but it is not able to download my citations into a file..

i did the same pubmed search using chrome and was able to download all citations into a file with an extension .nbib.. when i tried to import these references into qiqqa library using 'generic bibtex'. it imports these references but all the references are empty with no metadata...

kindly help with this situation...

regards, raza

so far i have managed to work around the problem by saving the pubmed search as a list of PMID (pubmed ID) in a text file and copy pasting this whole list into zotero which can get citations from PMID and it retrieved all of the references with a single click.. do we have a function in qiqqa that supports identifiers like PMID as this would be very useful..

Hi Ian,

please also guide whether (and how) qiqqa can help find available pdf for a vanilla reference.. i tried right click and selecting "download pdf from your web/intranet library' with no result..

many thanx in advance for your kind and patient support..

regards, raza

@Raza: In the "old days" we would draw up a reading list, and go to the library to get the articles out and photocopy them. Today we work mainly the other way around.

First we use Qiqqa to open a document. If we "want" the PDF file in our Qiqqa document library, we use the "Select text" (Ctrl-S) tool to highlight the title. We right click on the title and select "Search the web". We use the Scholar tab in preference to the PubMed tag, because PubMed does not provide a BibTeX import link.

Usually the first Scholar entry is what we are after. We click on Scholar's "Import into BibTeX" link. Click on "Convert to PDF" which will take the BibTeX data and add it to your BibTeX library --

thus working towards making your reading list complete!

I have not played around with .NBIBs or PMIDs, so cannot advise you there. Suggest your needs to the developer.

@Raza: The answer to your vanilla question is answered at https://getsatisfaction.com/qiqqa/topics/how_do_i_create_a_vanilla_reference_in_pubmed

@Julie: A (plain) "vanilla reference" is a reference with no attached PDF document. Perhaps you found it in an old dusty book that you were not allowed to scan, so you typed the details into your BibTeX library so that you could still reference it in your thesis-book.

Hi! The information that have been posted here, its almost a quick start to Qiqqa. I am thinking and copy the topics articulated in file (word and sharing it here) or to create a project (Qiqqa documentation) in researchgate. I have also access to an account to alfresco (document management system) that could be used for this, that is very suitable.

Sorry, this is just an idea following our posts and getting a materialization of what have been done here. Basic idea its to maintain the interaction here use it to start the "Qiqqa 101".

What do you think?

Thanks a lot Ian..

I was able to follow the Sniffer method mentioned in the link provided and actually download the article and associate it to the required reference.. so it is working but i wish there was an auto-pdf finder which would make my lazy life so much easier.. :)

thanks again.. regards

Hi Pedro,

It seems like a great idea.. .it would be really helpful for everyone to go back and find solution to the problems commonly faced..

@ Julie: qiqqa calls a reference without an associated pdf file a 'vanilla reference'.. :)

regards, raza

Ian do you think it is possible for me to 'sniff' multiple vanilla references for their respective pdf files in a way other than going through each reference one by one..

i imported into qiqqa library all the references that i had retrieved in zotero through their PMIDs... then i tried selecting all of the references and clicked on BibTeX Sniffer but it wouldn't work since the BibTeX Sniffer seems to be built to find references for pdf and not the other way round...

i'm only complaining as i just figured out there are 113 references to go through at about the same time as the excitement of finding this method and the caffeine levels came down... :)

please help..

regards, raza

@Julie: I am having the same problem trying to tag files with my preferred tags which i can use for later analysis but expedition seems to be ignoring my tags and creating themes most likely from autotags...

Exhibit A: tags vs expedition vs raza (File attached).. fluorescent green circle shows tags that i had allocated to a single file whereas blue circle shows the themes generated by expedition for this file.. my tags seems to have been altogether ignored...

@ Ian: i want to unselect 'Use Autotags?' checkbox shown by the red circle.. is there any way to do so while using the free version..

regards, raza

@Raza I am getting a different screen! It seems that there isn't enough information in the database to generate themes and your personal tags should be formulated another way.

Check page 43 of the manual @ http://www.qiqqa.com/Content/Client/The%20Qiqqa%20Manual.pdf and a mini manual @ http://www.donationcoder.com/forum/index.php?topic=34220.msg319664#msg319664

Regarding tags check this document https://getsatisfaction.com/qiqqa/topics/some_general_questions_about_tagging . I think that you should not use "vs" because you are trying to insert two keywords in the same tag, after you could interrogate Qiqqa for each tag using boolean operation e.g. finding both words "Lap AND Slice" or and any occurrence of "Lap OR Slice". I am newbie... but this seems a better way.

Thanks a lot Pedro.. i'll go through the manual and the links regarding this..

i understand how use of terms like 'vs' could create problems... my idea was to tag a study comparing two methods of treatment and i could organize similar articles according to my tags (for example levels of evidence, and methodology) then it will make sense to me as to how much weight i can give to each article when i analyze them and when i start writing-up...

i think i'll try something like lap_open instead of 'lap vs open' so that my tags stay distinct from the autotags as well as avoid confusion from double themes as it would be a single word...

i still would be very happy if i can generate themes according to the tags i have made after going through these documents since it would start making sense of a large amount of references... i can only find out if it works or not once i've tried it.. :)

thank you very much again for providing the useful links..

regards, raza

@Raza I added more documents to Qiqqa and now he his crunching the data, after I will try what you said! I want to see how some customizing of the tags its reflected... But seems to me that inserting individual tags in documents is difficult to be reflect in the overall themes, this demands more occurrences that depends on the size of the data analysis and keywords identified...

@ Pedro: i agree that maybe it will only start mattering once a manual tag has been set for many files, and probably then it will become a dominant theme...

one problem i am facing is to insert a single tag for multiple files.. although there is a method mentioned (Exhibit B: tags vs raza - file attached) but that probably is also a feature of the premium version..

i guess that means i will have to manually enter a tag for each of my references to see if it can generate a theme... even then, it can be done at the time of individual pdf review and will save a lot of time later when i am at the stage of analysis and manuscript writing...

i'll get back to you once i've tagged quite a few documents and report the results..

regards, raza

so this is how it went...

i manually added the tag 'raza' (talk about self promotion!!!) to all the pdf files (n=160) in my library... and i double checked using filters to make sure that every file had this tag manually added to it.. then i ran it through expedition...

my tag was no where among the themes... i then waited an hour and also sync'd the libraries so that the new information is assimilated by the software... but still my tag was not among the themes... so i'm kind of stuck and confused how to use the tags...

not sure if i should try the premium version to see if this works or not... but the best thing would be to go through the manuals Pedro has suggested before i do that..

regards, raza

@Julie: it is the same here with trying to figure out things in qiqqa that all other work is getting ignored.. But i am quite hopeful that once this is figured out, a lot of the work will become easier to handle...

Hi all,

I have gone through the links Pedro recommended and it seems i am either very high on caffeine or much more knowledgeable about tags... :)

here are a few things that i have learnt...

- tags can be used to organize your files in a heirarchy and can also be used to search them within the library (even if expedition does not pay attention to them for the moment..)

- an easy way suggested by a user was to start the tags with special characters which get sorted first and will always be seen on top in the list of tags..

for instance i may choose to use hashtag for the research topic i am looking at and ampersand for important themes, and asterisk for specific journals and underscore for study design.. this will allow me to look for a particular article related to specific topic and concerning themes with a particular study design... this is just an example and combination of tags can be used to search the library...

hope this helps... i will post screenshot of how i have organized my tags after i have properly tagged them this time..

regards, raza

Hi all,

so this is my progress so far..

i have designed my schema for organizing my files, this will help me later during the write-up phase as i can easily sort files in library according to the tags that i have created...

so here is the schema i will be following.. by putting special characters before each tag, these tags will stay on top and will be clustered together for me to use a combination of and/or/not to modify my search for articles..

# parent project

* important themes

> study population

^ exposure variable

} control group

@ outcome variable

$ study design

? comments on methodology

_ secondary outcomes

& journal

% year of publication

< institution

i have to go through each article to extract information regarding these tags and i have started doing this for projects that are in their write-up phase. i will get back once i am done with a project and hopefully share examples..

@Ian: we miss your calm and extremely helpful responses to our flustered and stupid queries...

regards, raza

thats great Julie, were you able to use the BibTeX sniffer after importing your books to generate the citation/metadata..

interestingly, i added a number of pdfs from my home computer onto qiqqa today and generating metadata using the BibTeX Sniffer actually worked quite well and smoothly.. i think the automatic ocr of the scanned pdf articles is a very helpful feature and i'm very happy using qiqqa for that matter..

regards, raza

Hi @Raza I think that I can clarify the search.

First if you try exclamation mark near the search box it gives you some guidelines related to search capabilities in Qiqqa, were you can use several operators as you described (AND, OR, NOT) for example:

(1) You want to search articles with the keywords bpm and framework in the title, you write in the search box: title:bpm and title:framework

(2) You want to search an article with the title bpm and the keyword processes in the abstract, you write: title: BPM AND abstract:"process modeling"

(3) You can do this with every BibTex fields available, I think that you can use a personalized field for cataloging an research area that you want to define, e.g. BPM

More information about search syntax is here: http://lucene.apache.org/core/2_9_4/queryparsersyntax.html and Qiqqa help.

Raza sorry, but I don't understand what you mean about some tags on top?

Hi @Julie sometimes I use also library of congress http://catalog.loc.gov the book are well classified and is very easy. I think this is the weak area of Qiqqa, in Zotero we can add entries with much more options, than to be limited to BibTex and scholar.

Good luck to your thesis!

Hi Pedro,

thank you for explaining the search technique, it makes sense now..

i'm sorry i have not been able to explain the idea about tags as well as i got it from a user who suggested it on get satisfaction.com.. there is usually a huge list of tags that you have to scroll down to find the ones that you want to search for or type in the box provided...

since the tags are arranged in alphabetical order, if the personalized tags start with special characters (e.g # or $ or ^), they will always be on top of that list.. that way the tags are all bunched together on top and it is easy to scroll down select them using a combination of 'and', 'or', 'not'..

i hope it makes sense now..

i agree, zotero gathers metadata for a large no. of pdf files with a single click.. and that is a very cool feature.. but still if the pdf only contains scanned images of printed text, then i think zotero cannot obtain metadata about it and you can't do anything the text of the document (like select for a web search or highlighting text etc.).. i find the ability to ocr these documents automatically in qiqqa a very useful fuction..

regards, raza

@Raza: (referring to caffeine level post above) Make a clear mental distinction between your Qiqqa Document Library, (consisting of PDF and DOC files, spread over your drive) and your Qiqqa BibTeX which contains the bibliographic details for those files, together with plain vanilla entries when files are not available or in a different medium (JPEGS, MP3s, MP4s, dusty book not scanned yet).

To let Qiqqa do a full text and theme search over your full Document Library, you must have a local copy of each PDF file on your hard drive, not just plain vanilla references. Otherwise your Internet connection will provide a too-sluggish response.

The whole purpose of Sniffer is to pull in (sniff for) the bibliographic information you need for your local files. Qiqqa Sniffer does for you what your subject librarian or departmental librarian used to do as a favour for you when you had forgotten to note down bibliographic details when you were photocopying!

Qiqqa Sniffer is provided for you to get your referencing right the first time, with the minimum amount of pain. The only way of doing this is to check each reference (plain vanilla or otherwise) one by one. Capture once; capture correct. {You do not want to have to check all your references when you are about to bind your thesis-book!}

@Raza: Your exhibit A refers.

Themes are not generated for one file; theme are generated for one library.

Basically there are two ways of approaching your Document Library:

Data driven ("What in the world is this set of files trying to tell me?") OR Theory driven ("We have this tradition of dividing up things this way, using this vocabulary: let us continue in that tradition")

Manual tags will be squeaky clean; Autotags will be fascinating.

Autotags are useful for those researchers who have no idea what their document library contains.

However, most often, researchers have read much of their documents as they have collected them. Manual tags are for those researchers who are approaching the library with a reasonable (preconceived or preliminary) view of what they anticipate or expect to find.

Autotags can be used to suggest some partitioning or tags that they cannot see, have overlooked and might like to consider adding to their manual tags. If you add the found auto tags to your manual list, you will have the best of both worlds with a manual tag Expedition.

The other way around, there is no way you can force a manual tag to become an autotag!

@Raza: (re: 'self-promotion' tags) Work the other way around: Search your whole library, with a filter tag=Raza. It should only list your 'self-promotion' files.

@Julie and Raza: Do not feel guilty about spending time using Qiqqa “instead of writing your thesis-book”, because using Qiqqa is in fact part of writing your thesis-book.

Thanks a lot Ian,

We were missing your patient comments on our queries... :)

it is making a lot of sense now.. and i was actually feeling quite comfortable going through all the files in the library with BibTeX sniffer one by one, knowing that if this is done correctly now, then i won't have to do it again..

i agree now that tags are quite useful, even autotags can give insight into something one maybe ignoring or overlooking... i am actually going through each pdf and tagging them according to my heirarchical schema so that i can quickly sort articles that i want from the library, as well as making comments on the annotation which will help me while writing the manuscript..

so far it seems to be heading in the right direction.. i'm grateful for everyone's contribution and support.. especially Ian for mentoring us through this phase..

regards, raza

@Raza: I have calmly slipped back on-line after slipping away to watch a good movie last night.

I am not sure why you would ever want to tag according to "&journal" (and publication date) as I never have read a paper which says "Nothing about XXX was found in the Journal JJJ before

YYYY" Maybe you will be the first to do so?

Remember you can always sort according to Journal= and even by year of publication and then apply your tag filters.

I find it best to tag according to why I want the document and

how I might use it or where I might use it (Paper-chapter name).

@Raza: To quickly enter tags, just type in the first 2 or 3 letters and those tags that you have already used with those letters adjacent will appear. Click on the correct one, and remember to press Enter (or

for multiple (mass) tags, press Ctrl-Enter.)!

This way you are less likely to suffer from tag proliferation, and are able to control the vocabulary more easily.

@Raza: Congratulations! "All" you have to do now is click on and put in at least 177 tags. Remember to press Enter each time after selection!

@Ian: I agree, it would sound quite naive to report in a manuscript that 'nothing about xxx was found in journal jjj before yyyy'.. and i really do not intend to have my name go down in the history of published literature as the first one to do so.. :)

i think i want to add those tags because this is how my memory works.. i'm always looking for 'what was that systematic review in 2008 bjs(british journal of surgery) about covering ileostomy for LAR?' or 'do you have a copy of the review article from last year's nejm about mesenteric ischemia?'... of course i can look them up on scholar, but if i have those article in my library on qiqqa.. i simply need to select the tags for the journal and the year and my search will become easier...

@Ian: please explain the multiple (mass) tags...

also i didn't get the reason for putting in so many tags for the untagged files... please explain..

regards, raza

@Julie: Actually i like this fact that there are less tags so that i can add only the one's that i am interested in and this will have an impact on the way themes are generated.. and i can add more documents as time goes on and this way can build the library and themes according to my 'point of view'.. i guess that is when i need autotags to see if there is any other way i need to look at the data..

its not actually the autotags that bother me.. i kind of get more irritated by some of the (manual) tags that are already present in file.. while most are relevant to the article, some of them are really irritating like 'humans', 'males', 'females', 'aged', 'adult', '80 and above' etc.. i try and resist my temptation to delete these but am really weak when it comes to irritating tags so i right click on the tag in the left hand pane (library view) and select 'rename or delete tag' and usually get rid of it from the whole library..

@Raza: I selected two documents by ticking their boxes, and "mass" tagged them. It worked. I am not sure if this is only in the Premium version.

Work the way your mind and memory works. You want your Scholar and PubMed interests in your portable Document Library. It's more efficient.

You should give at least one tag for each of your 177 untagged PDF file, otherwise you are relying on autotags.

I tag according to my perceived research interests. (I tag per paper title or chapter title or even per thought)

One document may catch my eye in more than one way or have two sections dealing with different, interesting stuff.

Multiple tags per document overcome the inadequacies of strict hierarchical classification.

I have tested up to 26 tags per document. However, in practice I seldom have even 3 tags per document.

How in the world did you get a stOOOpid '80 and above' as a manual tag? Did you import it from a previous document manager? You are wise to zap such stuff, unless you are into geriatrics!

As Julie says, you are the boss of your Document Library!

@Raza I think a tag should be interpreted as a keyword, you cannot combine a keyword with an operator, e.g. AND OR. This means if you want to search "80 and above" you use two tags "80" "Above". Combining the two tags is done by and operator "AND" "OR".

The question is made sense use the word "above", probably is very generic. Can you identify something with "above"?

TAGS are mechanisms for indexing documents and generating themes or to be used as personnel document classifiers. And the types of words more suited are probably verbs, substantives. Don't use conjunctions like "and" that is handled in the search!

Hi everyone.. sorry i got tied up with quite a lot of other stuff and couldn't come online...

yes these tags were already present in the files when i went through them... they couldn't have been assigned by qiqqa since they're not autotags.. i'm assuming they were either added at the when the pdf was created or may be included in the metadata by the search engine.. anyway, i do get irritated by these and get rid of them when i see them in a file..

@Ian: Thanks about the multiple tags advice... i experimented with it and found that i can do it for only upto 5 files in the free version.. even then it saves some time.. especially for tags that pertain to a particular topic... the other tags i still prefer to add while i am reading the pdf..

i think these tags sort of serve as notes for me.. and i can only say if they are useful for me or not after i have written a manuscript using them..

i also like the review pane which can actually mark articles that i have read or need to read again to sort a few things out..

regards, raza

@Julie: i think moving to premium was a good move.. i was also considering spending $4 to try the premium version for a month but for the moment i'll try to manage with the free version..

@Pedro: thank you for the tip.. i will keep that in mind when creating personal tags..

regards, raza

Hello guys.. just wanted to share a few tips i learnt regarding retrieval of metadata using qiqqa...

first of all, using bibtex, how to make sure that you've selected the right article..

please check exhibit A: BibTeX vs raza.

when you select the whole of the title from the article pdf, usually the first result is it, but sometimes it is not the right article.. a quick check to confirm that the bibtex belongs to the same article is, when you click on 'Import into Bibtex' using the scholar tab, all the author last names are highlighted in the article as mentioned in the red circle... if this doesn't happen, you should check again...

hope this is helpful..

regards, raza

I also found out how to use PMID from pubmed to retrieve metadata for a pdf file..

Exhibit B: PMID vs raza explains it..

you simply go into bibtex sniffer and change to 'PubMedXML' tab. in the search pane, type the PMID as shown and press enter.. the metadata will be retrieved and author names will be highlighted in the pdf file to confirm this is the right article..

regards, raza

Thanks for the tips Raza!

Pedro