DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature Dataset

In this paper, we propose and implement a systematic pipeline for the automatic classification of AI-related documents extracted from large-scale literature databases. This process results in the creation of an AI-related literature dataset named DeepDiveAI. The dataset construction pipeline integra...

Full description

Saved in:
Bibliographic Details
Main Authors: Xingzhou Liang, Xiaochen Zhou, Hui Zou, Yi Lu, Jingjing Qu
Format: Article
Language:English
Published: Tsinghua University Press 2025-06-01
Series:Journal of Social Computing
Subjects:
Online Access:https://www.sciopen.com/article/10.23919/JSC.2025.0007
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849426222135640064
author Xingzhou Liang
Xiaochen Zhou
Hui Zou
Yi Lu
Jingjing Qu
author_facet Xingzhou Liang
Xiaochen Zhou
Hui Zou
Yi Lu
Jingjing Qu
author_sort Xingzhou Liang
collection DOAJ
description In this paper, we propose and implement a systematic pipeline for the automatic classification of AI-related documents extracted from large-scale literature databases. This process results in the creation of an AI-related literature dataset named DeepDiveAI. The dataset construction pipeline integrates expert knowledge with the capabilities of advanced models, structured into two primary stages. In the first stage, expert-curated classification datasets are used to train a Long Short-Term Memory (LSTM) model, which performs coarse-grained classification of AI-related records from large-scale datasets. In the second stage, a large language model, specifically Qwen2.5 Plus, is employed to annotate a random 10% of the initially coarse set of classified AI-related records. These annotated records are subsequently used to train a Bidirectional Encoder Representations from Transformers (BERT) based binary classifier, further refining the coarse set to produce the final DeepDiveAI dataset. Evaluation results indicate that the proposed pipeline achieves both accuracy and efficiency in identifying AI-related literature from large-scale datasets.
format Article
id doaj-art-836bcc6172e2486fbc8ad30eae3489f7
institution Kabale University
issn 2688-5255
language English
publishDate 2025-06-01
publisher Tsinghua University Press
record_format Article
series Journal of Social Computing
spelling doaj-art-836bcc6172e2486fbc8ad30eae3489f72025-08-20T03:29:31ZengTsinghua University PressJournal of Social Computing2688-52552025-06-016215816910.23919/JSC.2025.0007DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature DatasetXingzhou Liang0Xiaochen Zhou1Hui Zou2Yi Lu3Jingjing Qu4Shanghai Artificial Intelligence Laboratory, Shanghai 200030, ChinaUniversity of Hong Kong, Hong Kong 999077, ChinaSchool of Cultural Heritage and Information Management, Shanghai University, Shanghai 200030, ChinaDepartment of Informatics, King’s College London, London, WC2R 2LS, UKShanghai Artificial Intelligence Laboratory, Shanghai 200030, ChinaIn this paper, we propose and implement a systematic pipeline for the automatic classification of AI-related documents extracted from large-scale literature databases. This process results in the creation of an AI-related literature dataset named DeepDiveAI. The dataset construction pipeline integrates expert knowledge with the capabilities of advanced models, structured into two primary stages. In the first stage, expert-curated classification datasets are used to train a Long Short-Term Memory (LSTM) model, which performs coarse-grained classification of AI-related records from large-scale datasets. In the second stage, a large language model, specifically Qwen2.5 Plus, is employed to annotate a random 10% of the initially coarse set of classified AI-related records. These annotated records are subsequently used to train a Bidirectional Encoder Representations from Transformers (BERT) based binary classifier, further refining the coarse set to produce the final DeepDiveAI dataset. Evaluation results indicate that the proposed pipeline achieves both accuracy and efficiency in identifying AI-related literature from large-scale datasets.https://www.sciopen.com/article/10.23919/JSC.2025.0007ai-related documenttext classificationlong short-term memory (lstm)bidirectional encoder representations from transformers (bert)large language model (llm)
spellingShingle Xingzhou Liang
Xiaochen Zhou
Hui Zou
Yi Lu
Jingjing Qu
DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature Dataset
Journal of Social Computing
ai-related document
text classification
long short-term memory (lstm)
bidirectional encoder representations from transformers (bert)
large language model (llm)
title DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature Dataset
title_full DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature Dataset
title_fullStr DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature Dataset
title_full_unstemmed DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature Dataset
title_short DeepDiveAI: Identifying AI-Related Documents in Large Scale Literature Dataset
title_sort deepdiveai identifying ai related documents in large scale literature dataset
topic ai-related document
text classification
long short-term memory (lstm)
bidirectional encoder representations from transformers (bert)
large language model (llm)
url https://www.sciopen.com/article/10.23919/JSC.2025.0007
work_keys_str_mv AT xingzhouliang deepdiveaiidentifyingairelateddocumentsinlargescaleliteraturedataset
AT xiaochenzhou deepdiveaiidentifyingairelateddocumentsinlargescaleliteraturedataset
AT huizou deepdiveaiidentifyingairelateddocumentsinlargescaleliteraturedataset
AT yilu deepdiveaiidentifyingairelateddocumentsinlargescaleliteraturedataset
AT jingjingqu deepdiveaiidentifyingairelateddocumentsinlargescaleliteraturedataset