Comparing selective masking methods for depression detection in social media (Record no. 19836)

MARC details
000 -LEADER
fixed length control field 03861nas a2200469 a 4500
005 - DATE AND TIME OF LATEST TRANSACTION
control field 20260817173419.0
008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION
fixed length control field 240229s2022 th u m tt 000 a eng d
035 ## - SYSTEM CONTROL NUMBER
System control number .b12421662
099 ## - LOCAL FREE-TEXT CALL NUMBER (OCLC)
Classification number AIT Thesis no.DSAI-22-04
100 0# - MAIN ENTRY--PERSONAL NAME
Personal name Chanapa Pananookooln
245 10 - TITLE STATEMENT
Title Comparing selective masking methods for depression detection in social media
260 ## - PUBLICATION, DISTRIBUTION, ETC.
Place of publication, distribution, etc. Pathum Thani, Thailand :
Name of publisher, distributor, etc. Asian Institute of Technology,
Date of publication, distribution, etc. 2022
300 ## - PHYSICAL DESCRIPTION
Extent 47 leaves :
Other physical details ill.
490 1# - SERIES STATEMENT
Series statement Thesis ;
Volume/sequential designation no. DSAI-22-04
500 ## - GENERAL NOTE
General note A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Data Science and Artificial Intelligence, School of Engineering and Technology
502 ## - DISSERTATION NOTE
Dissertation note Thesis (M. Sc.) - Asian Institute of Technology, 2022
520 ## - SUMMARY, ETC.
Summary, etc. Identifying those at risk for depression is a crucial issue in which social media provides an excellent platform for examining the linguistic patterns of depressed individuals. A significant challenge in a depression classification problem is ensuring that the predic tion model is not overly dependent on keywords, such that it fails to predict when key words are unavailable. One promising approach is masking, i.e., by masking important words selectively and asking the model to predict the masked words, the model is forced to learn the context rather than the keywords. This study evaluates seven masking tech niques, such as random masking, log-odds ratio, and the use of attention scores. In ad dition, whether to predict the masked words during pretraining or fine-tuning phase was also examined. Last, six class imbalance ratios were compared to determine the robust ness of the masked selection methods. Key findings demonstrated that selective masking generally outperforms random masking in terms of classification accuracy. In addition, the most accurate and robust models were identified. Our research also indicated that re constructing the masked words during the pre-training phase is more advantageous than during the fine-tuning phase. Further discussion and implications were made. This is the first study to comprehensively compare masking selection methods, which has broad implications for the field of depression classification and the general NLP. Our code can be found in: https://github.com/chanapapan/Depression-Detection
650 #0 - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical term or geographic name entry element Social media
General subdivision Data processing
650 #0 - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical term or geographic name entry element Machine learning
650 #0 - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical term or geographic name entry element Neural networks (Computer science)
700 0# - ADDED ENTRY--PERSONAL NAME
Personal name Chaklam Silpasuwanchai,
Relator term Chairperson
700 1# - ADDED ENTRY--PERSONAL NAME
Personal name Dailey, Matthew N.,
Relator term Examination Committee
700 0# - ADDED ENTRY--PERSONAL NAME
Personal name Mongkol Ekpanyapong,
Relator term Examination Committee
710 2# - ADDED ENTRY--CORPORATE NAME
Corporate name or jurisdiction name as entry element His Majesty the King{u2019}s Scholarships (Thailand),
Relator term Scholarship Donor
810 2# - SERIES ADDED ENTRY--CORPORATE NAME
Corporate name or jurisdiction name as entry element Asian Institute of Technology.
Title of a work Thesis ;
Volume/sequential designation no. DSAI-22-04
856 40 - ELECTRONIC LOCATION AND ACCESS
Materials specified Full-Text
Uniform Resource Identifier <a href="http://203.159.5.9/ait-thesis/Viewer/viewer.php?id=B20417">http://203.159.5.9/ait-thesis/Viewer/viewer.php?id=B20417</a>
907 ## - LOCAL DATA ELEMENT G, LDG (RLIN)
a .b12421662
b mnait
c a
902 ## - LOCAL DATA ELEMENT B, LDB (RLIN)
a 250428
998 ## - LOCAL CONTROL INFORMATION (RLIN)
Operator's initials, OID (RLIN) 0
Cataloger's initials, CIN (RLIN) 240313
First date, FD (RLIN) m
-- a
-- a
-- 0
945 ## - LOCAL PROCESSING INFORMATION (OCLC)
l mnarc
945 ## - LOCAL PROCESSING INFORMATION (OCLC)
l mnarc
945 ## - LOCAL PROCESSING INFORMATION (OCLC)
l mnait
945 ## - LOCAL PROCESSING INFORMATION (OCLC)
l mnait
942 ## - ADDED ENTRY ELEMENTS (KOHA)
Koha item type 67-Electronic Resource
942 ## - ADDED ENTRY ELEMENTS (KOHA)
Koha item type 40-Archives
942 ## - ADDED ENTRY ELEMENTS (KOHA)
Koha item type 20-AIT Publication
909 ## - LOCAL ITEMS USED
Barcode Barcode : -
CREATED CREATED : 2024-02-29
RECORD Id RECORD # : i13487875
LPATRON LPATRON : 0
LCHKIN LCHKIN : -
RENEWALS # RENEWALS : 0
-- # OVERDUE : 0
-- IUSE3 : 0
-- TOT CHKOUT : 0
-- TOT RENEW : 0
909 ## - LOCAL ITEMS USED
Barcode Barcode : 30050120899066
CREATED CREATED : 2025-07-03
RECORD Id RECORD # : i13537118
LPATRON LPATRON : 0
LCHKIN LCHKIN : -
RENEWALS # RENEWALS : 0
-- # OVERDUE : 0
-- IUSE3 : 0
-- TOT CHKOUT : 0
-- TOT RENEW : 0
909 ## - LOCAL ITEMS USED
Barcode Barcode : 30050120425151
CREATED CREATED : 2025-08-04
RECORD Id RECORD # : i1354293x
LPATRON LPATRON : 0
LCHKIN LCHKIN : -
RENEWALS # RENEWALS : 0
-- # OVERDUE : 0
-- IUSE3 : 0
-- TOT CHKOUT : 0
-- TOT RENEW : 0
909 ## - LOCAL ITEMS USED
Barcode Barcode : 30050120425144
CREATED CREATED : 2025-08-04
RECORD Id RECORD # : i13542941
LPATRON LPATRON : 0
LCHKIN LCHKIN : -
RENEWALS # RENEWALS : 0
-- # OVERDUE : 0
-- IUSE3 : 0
-- TOT CHKOUT : 0
-- TOT RENEW : 0
Holdings
Withdrawn status Lost status Damaged status Not for loan Home library Current library Shelving location Date acquired Total checkouts Full call number Date last seen Price effective from Koha item type Barcode Copy number Cost, normal purchase price
      Available for Loans Asian Institute of Technology Library Asian Institute of Technology Library Archives 17/08/2026   AIT Thesis no.DSAI-22-04 17/08/2026 17/08/2026 67-Electronic Resource      
      Available for Loans Asian Institute of Technology Library Asian Institute of Technology Library Archives 17/08/2026   AIT Thesis no.DSAI-22-04 17/08/2026 17/08/2026 40-Archives 30050120899066 1  
      Available for Loans Asian Institute of Technology Library Asian Institute of Technology Library AIT Publications 17/08/2026   AIT Thesis no.DSAI-22-04 17/08/2026 17/08/2026 20-AIT Publication 30050120425151 1 50.00
      Available for Loans Asian Institute of Technology Library Asian Institute of Technology Library AIT Publications 17/08/2026   AIT Thesis no.DSAI-22-04 17/08/2026 17/08/2026 20-AIT Publication 30050120425144 2 50.00
คัดลอกแล้ว!