A research of invisible errors in bibliographic data input and its impact on the quality and search accessibility
The research is devoted to a special class of errors in bibliographic data input into automated library information system, which is invisible to users, but affects the functioning of the electronic catalog system. The cause of the problem is the misuse of visually similar Latin characters where Cyr...
Saved in:
| Date: | 2021 |
|---|---|
| Main Author: | |
| Format: | Article |
| Language: | Ukrainian |
| Published: |
Інститут проблем реєстрації інформації НАН України
2021
|
| Subjects: | |
| Online Access: | http://drsp.ipri.kiev.ua/article/view/239252 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | Data Recording, Storage & Processing |
Institution
Data Recording, Storage & Processing| Summary: | The research is devoted to a special class of errors in bibliographic data input into automated library information system, which is invisible to users, but affects the functioning of the electronic catalog system. The cause of the problem is the misuse of visually similar Latin characters where Cyrillic characters should have been and vice versa.
The study is based on bibliographic information collected from 141 public libraries in Kyiv for the period from 1993 to 2021 (obtained from two sources). This allows fully explore the features of the problem, its prevalence and impact on the functioning of automated library information system and its OPAC module.
Attention is drawn to the text fields common in search and identification tasks — «Book Title», «Author», «Publisher».
The investigation provides one by information about: 1) the method of automatic error identification is applied; 2) prevalence of errors by type and their percentage in each source; 3) the impact of errors on the search; 4) the impact of errors on the search for duplicates; 5) distribution of errors by symbols; 6) errors and use of reference tables;
The research has shown that all characters with the same appearance are used incorrectly. The frequency of use of symbols differs significantly. There are many mistakes related to Cyrillic using in Roman numerals. Often some part of the number is written in Cyrillic and some part in Latin. But it affects comparison more than search.
The conclusions state that this class of errors affects the search accessibility of hundreds of book records in the libraries of Kyiv and provide suggestions for measures to eliminate and prevent errors in the future. Some records correspond to several real books, so there are thousands of real books in different libraries. The problem can be solved only with software using. Effective prevention is possible with the appropriate improvements of automated library information systems. Tabl.: 4. Fig.: 3. Refs: 8 titles. |
|---|