Deep Transfer Learning Based Intrusion Detection System for Electric Vehicular Networks
Open accessThe DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index.
Abstract
The Controller Area Network (CAN) bus works as an important protocol in the real-time In-Vehicle Network (IVN) systems for its simple, suitable, and robust architecture. The risk of IVN devices has still been insecure and vulnerable due to the complex data-intensive architectures which greatly increase the accessibility to unauthorized networks and the possibility of various types of cyberattacks. Therefore, the detection of cyberattacks in IVN devices has become a growing interest. With the rapid development of IVNs and evolving threat types, the traditional machine learning-based IDS has to update to cope with the security requirements of the current environment. Nowadays, the progression of deep learning, deep transfer learning, and its impactful outcome in several areas has guided as an effective solution for network intrusion detection. This manuscript proposes a deep transfer learning-based IDS model for IVN along with improved performance in comparison to several other existing models. The unique contributions include effective attribute selection which is best suited to identify malicious CAN messages and accurately detect the normal and abnormal activities, designing a deep transfer learning-based LeNet model, and evaluating considering real-world data. To this end, an extensive experimental performance evaluation has been conducted. The architecture along with empirical analyses shows that the proposed IDS greatly improves the detection accuracy over the mainstream machine learning, deep learning, and benchmark deep transfer learning models and has demonstrated better performance for real-time IVN security.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures



![Figure 4 Application of intrusion detection model for IVN traffic. The figure illustrates how the proposed IDS can detect the possible attack vectors within an in-vehicle Network. The car image is adopted from [47].](https://figures.cochat.ai/paper-figures/6b74f21f827844938fa930c141f7ab1f/deep-transfer-learning-based-intrusion-detection-system-for-electric-vehicular-figure-4.jpg)










Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
1 retracted
Lin, Y., Chen, C., Xiao, F., Avatefipour, O., Alsubhi, K., & Yunianta, A. (2020) · IEEE Transactions on Industry Applications · doi:10.1109/tia.2020.3009906 · Cited by 36 · cited 2×
“[32] proposed a DL-based intrusion detection system for CAN networks to detect DoS, fuzzing, and impersonation attacks particularly.”
BackgroundRetractedBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“This operation reduces drastically the number of trainable parameters in a deep model while enabling the use of a class activation map which allows an interpretation of the learned features [49].”
MethodChandola, V., Banerjee, A., & Kumar, V. (2009) · ACM Computing Surveys · doi:10.1145/1541880.1541882 · Cited by 11,687
Top 10 algorithms in data mining
Wu, X., Kumar, V., Ross Quinlan, J., Ghosh, J., Yang, Q., Motoda, H., McLachlan, G. J., Ng, A., Liu, B., Yu, P. S., et al. (2008) · Knowledge and Information Systems · doi:10.1007/s10115-007-0114-2 · Cited by 5,667 · cited 3×
“The DT algorithm is the most popular model that is used for the IDS domain, which is showed by the authors in [57,58].”
MethodA Survey of Data Mining and Machine Learning Methods for Cyber Security Intrusion Detection
Buczak, A. L., & Guven, E. (2016) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2015.2494502 · Cited by 3,332
Boosting for transfer learning
Dai, W., Yang, Q., Xue, G.-R., & Yu, Y. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273521 · Cited by 1,724
Integrating structured biological data by Kernel Maximum Mean Discrepancy
Borgwardt, K. M., Gretton, A., Rasch, M. J., Kriegel, H.-P., Schölkopf, B., & Smola, A. J. (2006) · Bioinformatics · doi:10.1093/bioinformatics/btl242 · Cited by 1,674 · cited 1×
“We used the Maximum Mean Discrepancy Equation (1) to calculate the difference between the source and target domains [48]:”
MethodTransfer learning using computational intelligence: A survey
Lu, J., Behbood, V., Hao, P., Zuo, H., Xue, S., & Zhang, G. (2015) · Knowledge-Based Systems · doi:10.1016/j.knosys.2015.01.010 · Cited by 968
Learning and evaluating classifiers under sample selection bias
Zadrozny, B. (2004) · Twenty-first international conference on Machine learning - ICML '04 · doi:10.1145/1015330.1015425 · Cited by 800 · cited 1×
“[38] proposed a model for intrusion detection that performs better in both labeled and unlabeled data.”
BackgroundIntrusion Detection System Using Deep Neural Network for In-Vehicle Network Security
Kang, M.-J., & Kang, J.-W. (2016) · PLOS ONE · doi:10.1371/journal.pone.0155781 · Cited by 678 · cited 2×
“[23] DL >95% High N/A [24] DL >85% Medium DoS, Command Injection, Malware [25] GAN >95% High DoS, Fuzzing, RPM, Gear attacks [26] DCAE >90% Medium DoS, Fuzzing, Impersonation [27] DL >95% High Spoofing, Replay [28] LSTM >80% Medium Spoofing, Replay, Flooding [29] ML >90% High DoS, Fuzzing, Spoofing [30] RNN >95% High…”
Background
Research works printed but not linked: 6
- 9. Robert Bosch GmbH CAN Specification. Postfach 1991, 2, 1-72.
- 16. Theissler, A.; Ulmer, D.; Dear, I. Interactive knowledge discovery in recordings from vehicle tests. In Proceedings of the FISITA World Automotive Congress, Budapest, Hungary, 30 May-4 June 2010; pp. 1-10.
- 43. Kang, H.; Kwak, B.; Lee, Y.H.; Lee, H.; Lee, H.; Kim, H.K. Car Hacking: Attack and Defense Challenge 2020 Dataset. IEEE Dataport 2021. [CrossRef]
- 54. Geron, A. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow; O'Reilly Media: Sebastopol, California, USA 2018.
- 55. Guyon, I. A scaling law for the Validation-set training-set valid ratio. AT T Bell Lab. 1997, 1, 1-11.
- 62. Goodfellow, I.; Bengio, Y.; Courville, A. Deep Learning; MIT Press: Cambridge, MA, USA, 2016.
Cited by
Items 101–153 of 153
Advances in Intelligent Vehicle Control
Carrillo, J. A. C. (2022) · Sensors · doi:10.3390/s22228622 · Cited by 3
“Smart sensors: development of advanced strategies using future smart sensor technology and intelligent sensor fusion for the measurement and estimation of vehicle states, tire and road conditions, situation awareness assessment, environment mapping, fault diagnosis, and driving conditions [3–6].”
Integration of Deep Learning Algorithms into VANET
Pandey, P., Sharma, D., & Agrawal, M. S. (2025) · Studies in computational intelligence · doi:10.1007/978-981-96-5190-0_3 · Cited by 2
Design and Implementation of Network Intrusion Detection System ( NIDS ) Based on Deep Learning
Wang, W., & Liu, Z. (2025) · Security and Privacy · doi:10.1002/spy2.70133 · Cited by 2
An intrusion detection system for Internet of Vehicles based on digital twin
Wang, Y., Qin, G., Sun, M., Liang, Y., Yang, X., Li, M., & Hu, C. (2025) · The Journal of Supercomputing · doi:10.1007/s11227-025-08044-6 · Cited by 2
Seeing the unseen: Intrusion attack detection in connected autonomous vehicles
Huraysi, T., Sun, R., Duan, H., Adu-Duodu, K., Ranjan, R., Wei, B., & Shah, T. (2025) · High-Confidence Computing · doi:10.1016/j.hcc.2025.100375 · Cited by 2
Naeem, H., Ullah, F., Mazhar, F., & Rafique, M. A. (2026) · International Journal of Critical Infrastructure Protection · doi:10.1016/j.ijcip.2026.100829 · Cited by 2
VAE-XGBoost: a hybrid intrusion detection system for next generation EV charging networks
Asim, M., Sair, F., Ishaq, M. H., Cengiz, K., Akleylek, S., & Ivković, N. (2026) · PeerJ Computer Science · doi:10.7717/peerj-cs.3506 · Cited by 2
A Hybrid Ensemble Learning Approach for Intrusion Detection in Internet of Vehicles
Muhammad, S., Mehmood, G., & Chrysostomou, C. (2025) · International Conference on Internet of Things: Systems, Management and Security (IOTSMS) · doi:10.1109/iotsms68530.2025.11408521 · Cited by 2
“These threats target various components of the interconnected vehicular ecosystem, allowing attackers to execute a wide range of malicious activities [2].”
Jo, H., & Kim, D. (2026) · Ad Hoc Networks · doi:10.1016/j.adhoc.2026.104224 · Cited by 2
Intrusion Detection System for Electric Vehicle Charging Station
Miskin, S. V., Chandaragi, P., & Wali, U. V. (2023) · International Conference on Mobile Networks and Wireless Communications (ICMNWC) · doi:10.1109/icmnwc60182.2023.10435897 · Cited by 2
“• SVM is well-suited for intrusion detection tasks because intrusion detection systems must be able to detect new types of intrusions that they have not been trained on [12].”
BackgroundVaishali Mishra, S. K. (2024) · Journal of Electrical Systems · doi:10.52783/jes.1535 · Cited by 2
RETRACTED: Optimized deep learning-based intrusion detection framework for vehicular network
Vishnukumar, R., & Ramaiah, M. (2023) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-233581 · Cited by 2
“[7]devisedIDSmodelbasedondeeptransfer learning.”
RetractedChowdary, P. S., & Vinod, D. (2022) · Advances in parallel computing · doi:10.3233/apc220092 · Cited by 2
Wang, Y., & Zhang, S. (2023) · International Conference on Artificial Intelligence and Autonomous Robot Systems (AIARS) · doi:10.1109/aiars59518.2023.00014 · Cited by 2
Atmakuru, A., Badii, A., Fatta, G. D., & Pujol, F. E. (2025) · doi:10.36227/techrxiv.175985766.66705364/v1 · Cited by 2
LeNet enabled intrusion detection system for iot- networks
Rakha, M. A., Khan, I. U., Hajjami, S. E., Hajjami, A. E., Nishat, F., & Kaushik, K. (2024) · AIP conference proceedings · doi:10.1063/5.0200343 · Cited by 1
Identifying Intruder in Artificial Intelligence of Things Using Digital Forensic Framework: A Review
Kulkarni, M. M. S., Dhotre, P. S., & Pathan, M. S. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-6678-9_43 · Cited by 1
Transferred-Learning Intrusion Detection System for Internet of Vehicles
Nguyen, T. H., & Cacho, J. R. F. (2025) · Advances in intelligent systems and computing · doi:10.1007/978-3-031-89063-5_64 · Cited by 1
An Advanced Fusion Neural Network Paradigm for Intelligent Cyber Security Anomaly Detection
Zhang, C. Y., Tu, X., Lin, X., Zhang, Y., & Hua, Z. (2025) · Electronics Letters · doi:10.1049/ell2.70429 · Cited by 2
Dong, H., & Kotenko, I. V. (2025) · International Conference on Innovations in Intelligent Systems and Applications (INISTA) · doi:10.1109/inista68122.2025.11249556 · Cited by 1
Intrusion Detection in IoT: Techniques, Issues and Approaches
Shrivastava, K., Tiwari, M., & Chakrabati, P. (2025) · doi:10.1109/cises66934.2025.11265312 · Cited by 1
Revisiting the Intrusion Detection in In-Vehicle Networks
Khan, M. A., Menouar, H., & Abdallah, M. M. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3628885 · Cited by 1
Reddy, P. S. K., U, N. T., Pavan, E., Ravish, R., & Honnavalli, P. B. (2025) · International Conference on Vehicular Technology and Transportation Systems (ICVTTS) · doi:10.1109/icvtts67119.2025.11296618 · Cited by 1
“Transfer learning is utilized to improve model generalization and performance, particularly when working with limited or domain-specific data [7].”
Machine Learning Based Intrusion Detection System for Internet of Vehicles
Ashraf, W., & Masoodi, F. S. (2026) · International Conference on Computing for Sustainable Global Development · doi:10.23919/indiacom70271.2026.11526666 · Cited by 1
“The accuracy range achieved by [15] using RF, [16] using ILSTM, [17] using DTL was 98% to 99%. our suggested Random Forest model, on the other hand, performed better, achieving 99.999% accuracy and high precision, recall, and F1-score.”
InfluentialGhadi, A., & Mohd, T. K. (2026) · Journal of the Brazilian Computer Society · doi:10.5753/jbcs.2026.6140 · Cited by 1
“…0.997; TNR: 0.999 [Hossain et al ., 2020a] Multiclass: Accuracy up to 99.99%, Recall ≈ 1.00, F1-score ≈ 0.999; Binary: Accuracy 100%, F1-score 1.00 [Mehedi et al ., 2021] Accuracy: 98.10%; Precision: 98.14%; Recall: 98.04%; F1-score: 97.83%; ROC-AUC: 95.42% [Narasimhan et al ., 2021] Accuracy:…”
InfluentialObfuscation-Based Mechanisms in Location-Based Privacy Protection
Wozny, B. E., Pan, L., & Anwar, A. (2022) · Studies in computational intelligence · doi:10.1007/978-3-030-96630-0_9 · Cited by 1
Self-citationIn-Vehicle Intrusion Detection based on Machine Learning
Saber, O., & Mazri, T. (2023) · International Conference on Wireless Networks and Mobile Communications (WINCOM) · doi:10.1109/wincom59760.2023.10322964 · Cited by 2
“…vehicle networks that use wireless communication technologies to enable V2X communication and in-vehicle networks that enable communication within the vehicle, i.e., communication between different vehicle components and systems, such as the transmission, brakes, and entertainment system [4].”
BackgroundSimulation of Network Security Situation Assessment Model Based on Machine Learning Algorithm
Fu, H. Y., Liu, M., & Zou, Y. (2023) · International Conference on Power, Electrical Engineering, Electronics and Control (PEEEC) · doi:10.1109/peeec60561.2023.00084 · Cited by 1
Kangralkar, Y. M., Pattar, S., & Bavalatti, S. I. (2025) · Authorea · doi:10.22541/au.174721730.09296483/v1 · Cited by 2
Design and implementation of scene visualization system based on dual engine structure
Liu, M., Luo, J., Zhu, J., & Kang, Y. (2023) · International Conference on Electronics, Electrical and Information Engineering (ICEEIE) · doi:10.1117/12.3008957
“By detecting the network, it can monitor internal attacks, external attacks and misoperation in real time, provide dynamic protection for the network, and greatly improve the security of the network 8 .”
BackgroundSecurity Situation Assessment Model of Big Data Network Based on K-means Clustering Algorithm
Zhang, X., & Huang, S. (2023) · International Conference on Power, Electrical Engineering, Electronics and Control (PEEEC) · doi:10.1109/peeec60561.2023.00131
Kangralkar, Y. M., Pattar, S., & Bavalatti, S. I. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4819129
A Convolution Neural Network and Genetic Algorithm-Based Abnormity Detection Model for Cyber Attacks
Wang, F., Yu, K., & Yang, L. (2024) · Journal of Circuits Systems and Computers · doi:10.1142/s0218126625501014
Vaishali Mishra, S. K. (2025) · Journal of Information Systems Engineering & Management · doi:10.52783/jisem.v10i13s.2151
Deep Learning Algorithms for Detecting Network Attacks—An Overview
Mythili, R., & Aneetha, A. S. (2025) · Wiley eBooks · doi:10.1002/9781394228522.ch15
Transfer Learning Model for Anomaly Detection in Data Streaming - Data Engineering Perspective
Suryadevara, G., Udayaraju, P., Pachipulusu, P., Gayathri, M., Sitharam, M., & Kumar, V. D. (2025) · International Conference on Machine Learning and Autonomous Systems (ICMLAS) · doi:10.1109/icmlas64557.2025.10968941
“Mehedi et al. (2021) implemented a Deep TL-based IDS method for IVN.”
Machine Learning-Based Intrusion Detection in Electric Vehicles with Hybrid Deep Neural Networks
Hussain, S., Sivaprasad, D., Sravani, G., Thulasi, A., & Venkata Jaya Sree, N. (2025) · International Conference on Data Science and Information System (ICDSIS) · doi:10.1109/icdsis65355.2025.11071213
“Consequently, real-time collection, analysis, and classification of a large number of CAN bus message records is required [16].”
Assessment of Red Tip Disease in Pineapple (Ananas comosus) on Peat Soil Using NDVI
Kassim, F. A. (2023) · AgroTech Food Science Technology and Environment · doi:10.53797/agrotech.v2i1.8.2023
Billah, M., Arafat, M. E., Islam, N., Imtiaz, A., Shatabda, S., Kaiser, M. S., & Orosz, T. (2025) · Network Modeling Analysis in Health Informatics and Bioinformatics · doi:10.1007/s13721-025-00564-8 · Cited by 1
“We develop our proposed DeepTranSeq technique based on deep learning inspired by Mehedi et al. (2021).”
Patil, S., Taneja, M., Raghu, N., Swain, B., Reddy, N. H., & Sheetal (2026) · SN Computer Science · doi:10.1007/s42979-026-04762-7
Hakim, L., Ghozi, W., Sani, R. R., Yassin, W. M., Handoko, L. B., & Rafrastara, F. A. (2025) · IEEE Asia-Pacific Conference on Geoscience, Electronics and Remote Sensing Technology (AGERS) · doi:10.1109/agers67633.2025.11446362
“…given that CAN was originally designed for conventional vehicles, it lacks built-in security mechanisms such as encryption and authentication [8][9] it makes IoV vulnerable to cyberattacks that can disrupt automated decision-making and trigger accidents, such as Spoofing, Sybil attack, DoS [10]…”
Jyothsna, V., Pasupuleti, M., Pappireddy, G. R., Palapati, D. K. R., Palthe, R. N., & Srilakshmi, K. (2026) · AIP conference proceedings · doi:10.1063/5.0312142
Logambigai, R., Yeddula, P. K. R., & Yarraguntla, S. R. (2026) · Contemporary Computing Innovations Conference (CCIC) · doi:10.1109/ccic68129.2026.11485993
Abdallah, H. F., Hassan, M., & Eltaief, A. (2026) · WSEAS TRANSACTIONS ON COMPUTER RESEARCH · doi:10.37394/232018.2026.14.34
Arunamary, S., & Sudhagar, G. (2026) · International Journal of Intelligent Transportation Systems Research · doi:10.1007/s13177-026-00680-3
“IDS are [critical in identify - ing malicious traffic in both IoT and IT systems and are continuously improved to enhance their effectiveness and accuracy] [12].”
A Robust Cyberattack Detection Scheme for Autonomous Vehicles in Controller Area Network
Yadav, J., Thapliyal, S., & Wazid, M. (2026) · Procedia Computer Science · doi:10.1016/j.procs.2026.06.010
Deep-Learning Intrusion Detection for Connected and Autonomous Vehicles
James, G. M. (2026) · International Journal of Information Technology Research Studies (IJITRS) · doi:10.63090/ijitrs/3139.3209.0030
Shana, T. B., Mondal, S., & Agarwal, M. (2026) · Cluster Computing · doi:10.1007/s10586-026-06329-w
“With respect to the CIC-IoT23 dataset, based on Table 16, the proposed model is compared with the studies [21] and [18 ] outperforms with an accuracy of 14.64% and 1.67%, with a precision of 13.3% and 1.6%, with a recall of 15.97% and 1.76%, and an F1-score of 14.63% and 1.94%, respec - tively.”
InfluentialSEPIV-IDS: A Structured Evaluation Pipeline for In-Vehicle Intrusion Detection Systems
Alves, L. M. D. S., Roque, A. D. S., & de Freitas, E. P. (2026) · International Conference on Computer, Information and Telecommunication Systems · doi:10.1109/cits70307.2026.11637341
“In [28], the authors explored the use of Transfer Learning with an adapted deep CNN (LeNet-type) for intrusion detection.”
Yang, C., Du, F., zhang, S., & 康, 文. (2026) · Engineering Research Express · doi:10.1088/2631-8695/ae9a2f
“…attention has been directed toward the development of efficient and reliable intrusion detection systems for in-vehicle networks, with the aim of identifying heterogeneous cyberattacks and mitigating the security risks encountered by intelligent vehicles in real-world operating environments [3].”
Gupta, V. P., Singh, S., & Tiwari, V. K. (2026) · International Conference on Green Technologies and Sustainable Solutions (GTSS) · doi:10.1109/gtss68743.2026.11679846
“Deep learning based IDS models encompass CNN, RNN/LSTM, autoencoders and deep transfer learning models which have shown to be highly accurate in identifying spoofing, replay and DoS attacks to networks in in-the-vehicle and vehicular environments [6][7].”
Hassan, S. R., Din, S., & Mohmand, M. I. (2026) · Sensors · doi:10.3390/s26196211
“Mehedi et al. [38] presented a deep transfer learning-based IDS for vehicular networks, demonstrating how to leverage pre-trained models to enhance detection accuracy while reducing training overhead.”
Shukla, S. G., & Bhatnagar, M. (2026) · International journal of Computer Networks & Communications · doi:10.5121/ijcnc.2026.18502
Metadata
- Authors
- Mehedi, S. T., Anwar, A., Rahman, Z., & Ahmed, K.
- Affiliations
- Mawlana Bhashani Science and Technology University (BD)
- Deakin University (AU)
- Corresponding author
- Anwar, Adnan
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2021-07-11
- Volume
- 21
- Issue
- 14
- Pages
- 4736
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC8309471
- OpenAlex
- W3178719887
- Semantic Scholar
- a223b966b30f3a62da8f2cb13dc17d4f05d88560
- Cited by
- 150
- References
- 54
- Citations by year
- 2026: 19 · 2025: 47 · 2024: 31 · 2023: 29 · 2022: 24
- FWCI
- 9.40
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Figshare (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Immune Systems Applications
- Keywords
- intrusion detection systems, CAN bus, automotive network security, deep transfer learning, LeNet, feature selection, real-time network security
- MeSH terms
- Machine Learning, Electricity, Neural Networks, Computer
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.