=Paper=
{{Paper
|id=Vol-2870/paper41
|storemode=property
|title=Electronic System "All-Ukrainian Toloka Archival Card Index": Structure, Tools, Prospects of Development
|pdfUrl=https://ceur-ws.org/Vol-2870/paper41.pdf
|volume=Vol-2870
|authors=Oksana Tyshchenko,Vladyslav Tyshchenko
|dblpUrl=https://dblp.org/rec/conf/colins/TyshchenkoT21
}}
==Electronic System "All-Ukrainian Toloka Archival Card Index": Structure, Tools, Prospects of Development==
https://ceur-ws.org/Vol-2870/paper41.pdf
Electronic System «All-Ukrainian Toloka Archival Card Index»:
Structure, Tools, Prospects of Development
Oksana Tyshchenkoa and Vladyslav Tyshchenkob
a
Institute of Ukrainian Language of the National Academy of Sciences of Ukraine, M. Hrushevsky str. 4, Kyiv,
01001 Ukraine
b
National Pedagogical Drahomanov University, Pyrohov str. 9, Kyiv, 02000, Ukraine
Abstract
The article covered the principles and tools of collective recognition of manuscripts of the
Archival Card Index (ACI) – lexical and phraseological materials of the Commission for
compiling the Dictionary of the living Ukrainian language of the All-Ukrainian Academy of
Sciences. In 2018, the Institute of the Ukrainian Language of the National Academy of
Sciences of Ukraine created an electronic system «Archival Card Index» (ESACI) – digital
format of ACI. ACI (350 thousand units) has got a great importance in the context of the
cultural and national revival in Ukraine in the early 20th century, as it plays an important role
in the development of the Ukrainian language, the theory and practice of Ukrainian studies in
the 20th – early 21th century. The ACI fragment (3000 units) was recognized manually: the
texts were entered into the ESACI according to the fields of the microstructure of the card.
Such recognition requires considerable the effort and the time, so the platform «All-Ukrainian
Toloka Archival Card Index» (AUTACI) has been created on the ACI website, which provides
unlimited simultaneous online participation of volunteers for manual card recognition.
Collective access to the collection of the transcribed documents is accompanied by instructions
and samples of execution. The form for filling in the card is simplified in contrast to the form
in the ESACI, as we plan to involve non-specialists in the work. Access to the AUTACI is
possible after registration and has no time limits. In the future, we plan to use it to create tools
for future verification of ACI texts, were automatically recognized by the Transkribus
software, and for the partition linguistic information in the appropriate fields.
Keywords 1
Archival Card Index (ACI); Electronic System «Archival Card Index» (ESACI); open platform
«All-Ukrainian Toloka Archival Card Index» (AUTACI); Ukrainian Lexicography; Manual
Handwriting Texts Recognition; Lexicographic Toloka (Crowdsourcing)
1. Introduction
Crowdsourcing involves obtaining work, information, or opinions from a large group of people who
submit their data via the Internet, social media, and Smartphone apps while tapping into people with
different skills or thoughts from all over the world. Participants work on a paid or free basis as
volunteers. Crowdsourcing is becoming a popular method to raise capital for special projects, taps into
the shared interests of a group. It usually involves taking a large job and breaking it into many smaller
jobs that a crowd of people can work on separately, usually sourced via the Internet, it contributes to
save time and money [9].
Problems in the field of open innovation have been studied since the late 80's of the 20th century.
This is especially determined by the sources of innovation and their dynamics [20], finding ways to
solving problems, in particular, resolve them into local tasks [5; 8; 19]. Today the specialized interest
COLINS-2021: 5th International Conference on Computational Linguistics and Intelligent Systems, April 22–23, 2021, Kharkiv, Ukraine
EMAIL: tom-73@ukr.net (O. Tyshchenko); vvtyshchenko@gmail.com (V. Tyshchenko)
ORCID: [0000-0002-5709-1252] (O. Tyshchenko); [0000-0003-2449-4722] (V. Tyshchenko)
©️ 2021 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
of researchers of open innovations is concentrated mainly in the field of business, in particular in the
marketing [6; 7]. These are the benefits of engaging consumers in the product development and support,
and also related activities. The such technology systems can improve the customer experience and can
help companies to improve their innovation and the customer's relationship management
capabilities [12]. Such models of open platforms began to be used (and quite effectively) in the field of
public administration [10].
Open platforms are also used in scientific research, particularly in philological research. Thus, in the
OpenCorpora project, scientists try to involve native speakers who do not have special linguistic
knowledge in the annotation works. To do this, a method of organizing processes to support of high
quality annotations has been created [4]. Another example: the merger of synsets (synonymous set) to
integrate two lexicographic resources (RussNet and YARN) was implemented by the authors,
particularly, through a combination of experts and crowdsourcing approaches. The developers
emphasize that the crowdsourcing methodology is a new and relevant area of researches in many
areas [3]. Crowdsourcing is used by the Transkribus automatic text recognition system team to solve
such problems. For faster processing of texts, collective access to the collection of transcribed
documents is provided in the Transkribus Web interface – a lightweight, convenient and easy to use
version of Transkribus [14]. In Ukraine collective contributions, mainly from students, primarily enrich
the corpus resources: GRAK (General Regional Annotated Corpus of the Ukrainian Language)
regularly attracts volunteers to increase and diversify the text, particularly, during the distance practical
training in applied linguistics [13].
In the same way, everyone can join our common cause – the recognition of handwritten cards of the
Archival Card Index (ACI). The stamp «ARCHIVAL», which marked the cards in the 50's of the 20th
century, gave a conditional name to the Card Index itself. ACI was compiled in the 20s of the 20th
century for the work of the Commission of the Dictionary of the Living Ukrainian Language Compiling
of the All-Ukrainian Academy of Sciences. Several iconic Ukrainian dictionaries (first of all, «Russian
Ukrainian Dictionary». V. I–III, Ch. edit A.Yu. Krymsky (I–II V.), S.O. Yefremov (III–IV V.), Kyiv
1924–1933) of the Golden Age of the Ukrainian lexicographic were compiled on the basis of this card
index. In the 1930's, after the «purity» from «national junk» (the decline of Ukrainization and the
beginning of mass repressions among all segments of the Ukrainian population, including the
humanitarian sphere, from 1929 and, especially, in 1933-1937), the ACI was temporarily abandoned.
In the 1950's it was combined with millions of cards of the new-created Lexical Card Index (hereinafter
– LCI) of the O. Potebnya Institute of Linguistics of the UkSSR’s Academy of Sciences. Later, the LCI
(included the forgotten ACI) inherited the Institute of the Ukrainian Language of the National Academy
of Sciences of Ukraine. ACI are materials with a lost and forgotten history. It should be updated in a
modern field of question to find answers to common problems of restoring and strengthen of the
Ukrainian language identity [17].
ACI contains two types of the cards: monolingual cards (with one title word – Ukrainian; it is 1/3 of
the ACI) and bilingual cards (translated – Ukrainian-Russian, Russian-Ukrainian; 2/3 of ACI). These
materials were prepared mainly for translated dictionaries. Headwords are usually accompanied by a
quote and by a passport of the source. First, they are the working materials of the lexicographer, which
shows the dynamics of scientific research. Secondly, the valuable linguistic facts recorded both in the
title words and in the quotations, which can be explored from the point of view of both the individual
and the collective linguistic creativity (Figure 1).
These are rare materials for linguistic researches in many areas, they need to be preserved and
involved in modern linguistic processes. Therefore, it is logical to digitize the ACI and make it available
on the Internet. The preparatory stage of ACI digitization became possible thanks to the Toloka (a
toloka – ukr. толока ‘crowdsourcing’): hundreds of volunteers joined the all-Ukrainian action
«Preserve the Archival Lexical Card Index!» and manually processed about 6 million LC cards in order
to choose 350 thousand cards with the «ARCHIVAL» stamp.
2. Scientific Novelty
In 2018, the Institute of the Ukrainian Language of the National Academy of Sciences of Ukraine
created an Electronic System «Archival Card Index» (ESACI) – digital format of ACI (Figure 2).
Figure 1: Cards of Archival Card Index
A fragment of the ACI (about 3000 cards) has already been recognized, that is the texts of the cards
have been manually transcribed. If the card has not recognized yet, it would only be viewed as an image.
So, today the search for a given word for unrecognized cards (and this array of ACI predominates) is
impossible (Figure 3).
The Card recognition is the entering of text in the appropriate fields that reflects the microstructure
of the card, e.g., the headings and additional words or descriptive constructions: rus. title: свернуться,
rus. additional: свернувшийся – ukr. title: ске́плений (Figure 4). We could also sequentially record
all corrections in the cards or later added items, e.g., to стемнеть added повечереть. Such inserts
we fix as additional to the title word of the unit, in this case – a synonym (in detail the structure of the
ESACI is described in [15]).
Figure 2: Electronic System «Archival Card Index»: web interface (2018–2021)
The manual recognition requires considerable effort and time. From 350,000 ACI cards, about 3,000
have been transcribed. The first fragment of the Archival Card Index was recognized within two months
by four project participants in the Electronic System «Archival Card Index» (ESACI) in the offline
mode, before the launch of the system «Toloka» [15]. Therefore, we see the point in using technologies
that speed up this process. For example, there is a System of Automatic Text Recognition –
Transkribus [14] (we have prepared the article «Synopsis: text, context, media» about machine
recognition of handwritten text and preparation of archival data cards for it in «Ukrainian Language»
journal (publication is expected in May, 2021). Of course, the description will need to check the
recognition of automatic text and to enter information in the appropriate fields of the ACI system (to
automate the work with the materials of the card index).
Figure 3: Box №8: the images of scanned ACI cards, which are contained in the same order in the box
№8, the bookcase №1 on paper form
Figure 4: The ACI Card, manually recognized in ESACI
We can speed up the correction of cards within the framework of our new project «All-Ukrainian
Toloka Archival Card Index» (AUTACI) – an online platform on the ACI website for manual
recognition of card texts by everyone interested [1] (Figure 5). Interested persons can register and take
part in the manually recognition of card texts of the ACI within the Lexicographical Toloka. As already
mentioned, the collective addition of volunteers to the affairs of the ACI – the selection and sorting of
paper cards – has become traditional.
3. Basic Information about the «All-Ukrainian Toloka Archival Card Index»
(AUTACI)
Consider the structure of AUTACI, its content, tools. Toloka is available at http://work.iul-
nasu.org.ua/web/. The main page (Figure 5) contains a description of this toolkit: it states its purpose,
gives instructions for work and prospects for the application of results. We present these aspects step
by step.
ALL-UKRAINIAN TOLOKA.
What? Recognition of the Archival Lexical Card Index:
• card marking for the input language;
• entering the text of the card according to the fields (elements of the card microstructure).
How? There are six basic steps you need to take.
Step 1. Register on the site: login, password (received from the administrator, «Login» button).
Step 2. Select a bookcase (Figure 6), a box (Figure 7), a card (Figure 8). There will be two
bookcases in total. The structure of the electronic file match to its real state in paper form.
Figure 5: Section «All-Ukrainian Toloka» on website ACI: The Manual
During operation, the downloaded cards receive the status displayed on the interface with color:
blank (gray), completed (green), uncompleted (yellow), in operation (blue) (Figure 8). By selecting the
desired card (usually blank), the operator recognizes it by filling in the specified fields. If during the
transcription there are questions and the card needs to be finalized, the operator will select the Difficult
check box and the card will receive the status uncompleted. If the transcription is successful, he will
select Done → Save and the card will receive the status completed. Now the site has a simple filter to
go to the next card: next / previous (all in a row); next / previous completed; the next / previous
uncompleted (Figure 9). Subsequently, we plan to distinguish between the transition to the next card
for the administrator and the operator (see below).
Step 3. Enter the following text in the formula for recognizing ACI cards in the appropriate
fields (Figure 9).
Figure 6: Select the bookcase
Figure 7: Select the box
Card type – determine the existing input language that is written with the first (or only) title
word. We distinguish the following types: Russian-Ukrainian, Ukrainian-Russian, Ukrainian,
Russian, it is impossible to determine (e.g., in the case of мир / мир). In Figure 5 – Ukrainian-
Russian card.
Ukrainian title word: Сідати.
Russian title word: Садиться.
Quote – enter the quote-illustration: Семен зрадів так прохав брата сідати....
Source – enter the source (passport) of the quote as written on the card: Коцюб. І. 120. Ціпов.
Author – decipher the source and write the name / pseudo / title of the publication as a
condition of the name of the source to form a convenient search list: Франко, Українка, Федькович,
Кримський and others. (regardless of how it is written on the card). The author of the card in Figure 5
– Коцюбинський.
We focus the operators' attention on the important conditions of the ACI cards transcription:
accurately and truthfully reproduce the text of the card: with all elements, signs, abbreviations, as
well as errors (if any), format (italics, underlines, strikethroughs and inserts, uppercase / lowercase).
We remember the pre-reform elements in the Russian part of some cards: і, ъ, ѣ: семь лѣтъ,
Семилѣтній, about the older Ukrainian spellings, dialectal variants: лїс instead of the modern
normative ліс, життє instead of життя, etc. Technically, this feature is provided by a panel in
each text input field, which contains special characters and means of a text editor (Figure 9).
If you have got the difficulty decoding text, when it is heavily to understand what is written, then
you should use the following tips:
CTRL+ – the card image can be enlarged.
DIDN'T HELP? Google for help: you look for the title of the work by the author or quote, the
author's name – by the quote, etc. It is possible to copy and paste the text in the appropriate field.
DIDN'T SAVE YOU? Denote the unclear fragment by dots in double square brackets [[...]].
DO YOU DOUBT, that everything was done correctly? Press the «Difficult» button – the card
receives the status uncompleted (yellow), it will be checked and completed later.
Figure 8: Select the card: blank (gray), completed (green), uncompleted (yellow), in operation (blue)
Figure 9: The form for ACI card recognition
Step 4. Completed – press when you are sure that everything was done correctly. The card receives
the status completed (green).
Step 5. Save – save the recognized card.
Step 6. Next – choose a new card: next / previous (all in a row); next / previous completed; next
/ previous uncompleted.
We remind you that the structure of the form for recognition in AUTACI is simpler than the structure
of the card in ESACI. If the operator is occurred a very difficult card that contains a lot of different
information not provided by the form (e.g., additional words and sources, grammatical notes, more
difficult structure, many edits), he will skip it and goes to the next card, where everything is clear and
simple. Difficult cards are for the next level of work in ESACI, more professional. In the AUTACI,
they will retain the status of blank.
Access to the site is possible for two types of users: administrator and operator, respectively, with
different levels of rights.
1. The administrator may: register operators; add cards; monitor the status of work performed –
recognize and check the cards recognized by other participants, monitor statistics, history of changes in
each card (Figures 10, 11, 12, 13).
2. The operator may: enter information into the recognition form; check the work done, make
changes to his cards; review the work done by other participants (Figures 6, 7, 8, 9).
Figure 10: History of changes
Figure 11: Administrator’s excess: Add Card
4. Conclusion and Future Work
Subsequently, the interface will be improved. The pages of ESACI and Toloka (AUTACI) will have
mutual hyperlinks. We will also improve the pass to the next page for the administrator and the operator
to check the selected type of cards (completed and uncompleted) and general monitoring of the work.
Also will be created new filters:
for the administrator – transition with the right to view and make changes to the cards: a) for
the operator: Petrenko, Ivanenko, Sydorenko, etc. or all in a row; b) for readiness: completed;
uncompleted; blank; all in a row.
for the operators – transition with the right to work with cards with different degree of
completion: a) to view the cards (own / other operator); b) to make changes to the cards (only their
own).
Figure 12: Administrator’s excess: Users (Operators) Management
Figure 13: Administrator’s excess: Statistics
There is no doubt that the trend towards «open innovation» has revived interest in using external
sources of innovation. Different societies, institutions, firms purposefully open their models for
connection of internal and external ideas, for joint creation of values with their partners and users.
Internet platforms for Toloka (Crowdsourcing) and Co-creation have changed the way open innovation
is introduced. They have provided new ways to work together to solve problems and create values.
Toloka cause considerable interest in the community and gives real results [11].
What will happen from this? We emphasize that ACI and its component AUTACI are only a tool
for studying the lexicographic values of the Archival Card Index. Creating a text version of the ACI
will give everyone the opportunity to work with words, quotes, sources. It will let understand what
language material could be fill the 4th repressed volume of the «Russian-Ukrainian Dictionary» 1924-
1933, ed. A.Yu. Krymsky and S.O. Yefremov and other dictionaries written but not published up at that
time. This capability will be achieved through tools such as:
fast and convenient search;
sorting for different filters (by word, phrase, author, work and many others);
creation of a register (Russian, Ukrainian, translated Russian / Ukrainian, Ukrainian / Russian);
review of edits in the cards and changes in the sources of language material, understanding of
the dynamics of the lexicographer's thinking;
observation about finding the necessary match to the word and much more.
This will enrich the tools and will expand the language base of linguistic research, will contribute to
the creation of modern dictionaries, guides, grammars. In general, the ACI and the electronic resources
were created for its processing will play an important role in renewal and preserving the identity of the
Ukrainian language. Therefore, the actual scientific studies of ACI in linguistic and lingua-cultural
optics are promising and important.
5. Acknowledgements
We thank the volunteers who joined the Toloka-2018 – choosing the cards of the Archival Card
Index: students and teachers of the National Pedagogical Dragomanov University, Kyiv B. Grinchenko
University, Kyiv National T. Shevchenko University, speech therapists and musicians, publishers and
photographers, philologists and aircraft builders, teachers and school students from Kyiv, Lutsk,
Kherson, Boryspil. We express our gratitude for the help to Prof. P. Hrytsenko and Prof. E. Karpilovska
(Institute of the Ukrainian Language of the National Academy of Sciences of Ukraine), to Assoc.
Prof. V. Starko (Ukrainian Catholic University).
The project Electronic System «Archival Card Index» (ESACI) – digital format of ACI was
supported by the Ukrainian Cultural Foundation, grant № 1109, September – November 2018. The
work was attended by: PhD in Philology O. Tyshchenko (project manager) and DSc in Philology
L. Kysliuk (researchers of the Institute of Ukrainian Language NAS of Ukraine); PhD in Philology
Yu. Pozdran (senior lecturer at Vinnytsia National Technical University) and Yu. Vozniuk (graduate
student of the Institute of the Ukrainian Language of the NAS of Ukraine). The electronic system was
implemented with the participation of programmers led by M. Tkachenko.
6. References
[1] All-Ukrainian Toloka: Archival Card Index, 2020, URL: http://work.iul-nasu.org.ua
[2] Archival Card Index, 2018–2021, URL: https://ak.iul-nasu.org.ua
[3] I. Azarova, P. Braslavsky, V. Zakharov, Yu. Kiselev, D. Ustalov, M. Khokhlova, RussNet and
YARN. In: Structural and Applied Linguistics, vol. 12, St. Petersburg, 2019, pp. 34–52.
[4] V. Bocharov, S. Alexeeva, D. Granovsky, E. Protopopova, M. Stepanova, & A. Surikov.
Crowdsourcing morphological annotation. In: Computational linguistics and intelligent
technologies, Bekasovo, 2013, pp. 109–114, URL: http://www.dialog-
21.ru/digests/dialog2013/materials/pdf/BocharovVV.pdf
[5] H. W. Chesbrough, (Ed.), Open innovation. The new imperative for creating and profiting from
technology. Harvard Business School Press, Boston, 2006, 227 p.
[6] H. W. Chesbrough, W. Vanhaverbeke, J. West (Еds.) Open innovation: Researching a new
paradigm, Oxford Univ. Press, Oxford, 2006, pp. 1–12. URL:
http://scholar.google.com/scholar_lookup?&author=H.W..%20Chesbrough&pages=1-
12&publication_year=2006.
[7] K. Brockhoff, Customers’ perspectives of involvement in new product development. Int.
J. Technology Management 5/6, 2003.
[8] H. Chesbrough, A better way to innovate. Harvard Business Review, vol. 81(7):12–3, Boston,
2003, 115 p.
[9] General Regionally Annotated Corpus of the Ukrainian Language (GRAC). M. Shvedova, R. Von
Waldenfels, S.Yarygin, M. Kruk, A.Rysin, V. Starko, M.Wozniak, Kyiv– Oslo–Yen, 2017–2019,
URL: https://www.uacorpus.org
[10] G. Koch, J. Füller, S. Brunswicker: Online Crowdsourcing in the Public Sector: How to Design
Open Government Platforms. In: International Conference on Online Communities and Social
Computing OCSC 2011: Online Communities and Social Computing, 2011, pp. 203–212, URL:
https://link.springer.com/chapter/10.1007/978-3-642-21796-8_22
[11] H. Marshall, Crowdsourcing. Investopedia. 2019, URL: https://www.investopedia.com/terms/c/
crowdsourcing.asp
[12] S. Nambisan, P. Nambisan: How to profit from a better virtual customer environment. In: MIT
Sloan Management Review, vol. 49, 2008, pp. 53–61, URL:
https://sloanreview.mit.edu/article/how-to-profit-from-a-better-virtual-customer-environment/
last accessed 2021/01/31
[13] Students of the Institute of Philology joined the development of the General regionally annotated
corpus of the Ukrainian language. Institute of Philology of Kyiv B. Hrinchenko University. URL:
https://if.kubg.edu.ua/prouniversitet/podii/1418-studenty-instytutu-filolohii-doluchylysia-do-
rozrobky-heneralnoho-rehionalno-anotovanoho-korpusu-ukrainskoi-movy.html
[14] Transkribus, 2021, URL: https://readcoop.eu/transkribus/ Transkribus | Handwritten Text
Recognition | READ COOP
[15] O. Tyshchenko, Archival card index of the Ukrainian language in digital format: from a language
monument to modern lexicographic tools. In: Rocznik Slawistyczny. vol. LXIX, Wrocław, 2020,
pp. 185–197
[16] O. Tyshchenko, Electronic lexical card index: the way to create modern vocabulary tools. In:
Ukrainian language, 2, 2019, pp 37-52
[17] O. Tyshchenko, The archival card index as the lexical and illustrative base of «Russian-Ukrainian
dictionary» ed. A. Krymsky and S. Yefremov part 1. Lexical card index: history of creation and
repression; II. Micro- and macrostructure of archival lexical card index. In: Ukrainian language, 2,
2016, pp. 44–71; 3, 2016, pp. 57–78
[18] O. Tyshchenko, V. Tyshchenko, Metadata of the Linguistic Sourcesin Lexicographic Electronic
Tool. In: Computational Linguistics and Intelligent Systems. Proceedingsof the 4th International
Conferenceon Computational Linguistics and Intelligent Systems (COLINS). vol. I: Main
Conference (Lviv, Ukraine, April 23–24), 2020, URL: http://ceur-ws.org/Vol-2604/paper24.pdf
[19] E. von Hippel, Sticky information and the locus of problem solving. Implications for innovation.
In: Management Science, vol. 4, 1994, pp. 429–439
[20] E. von Hippel, The sources of innovation. Oxford University Press, New York, 1988, 221 p
[21] Material for the SANU Dictionary. Institute for Serbian Language SANU, 2018, URL:
http://www.isj.sanu.ac.rs/2018/10/04/gradja-za-recnik-sanu-blago-koje-treba-sacuvati/