Enhancing Learning Path Recommendation via Multi-task Learning

Afsana Nasrin; Lijun Qian; Pamela Obiomon; Xishuang Dong

REVIEW 4 major objections 6 minor 30 references

Enhancing Learning Path Recommendation via Multi-task Learning

T0 review · 4 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A shared-LSTM multi-task model with a non-repeat penalty beats six baselines for learning path recommendation on ASSIST09.

desk verdict A reasonable multi-task extension, but the core mechanism is not implementable from the text and the empirical claims are under-supported. read the letter →

arxiv 2507.05295 v1 pith:X4NLXUCK submitted 2025-07-05 cs.IR cs.AIcs.CYcs.LG

Afsana Nasrin , Lijun Qian , Pamela Obiomon , Xishuang Dong This is my paper · ORCID

classification cs.IRcs.AIcs.CYcs.LG

keywords learningpathrecommendationmulti-taskLSTMdeepknowledgetracingsequence-to-sequencepredictionnon-repeatlossASSIST09

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that learning path recommendation improves when it is trained jointly with deep knowledge tracing through a shared LSTM encoder, and when a penalty discourages repeated items in the recommended path. The authors reframe recommendation as sequence-to-sequence prediction: given a student's recent exercise history, the model generates the next several exercises. On the ASSIST09 dataset, the combined model reports the best scores on every metric, with accuracy 0.3489, F1 0.3241, precision 0.3202, and recall 0.3489. If correct, this means a modest multi-task architecture with an explicit diversity penalty can outperform more elaborate RNN, LSTM, and attention-based baselines on next-item learning path prediction.

What carries the argument

The load-bearing object is a shared LSTM encoder followed by two task-specific LSTM heads: one for next-item path prediction trained with cross-entropy and one for deep knowledge tracing trained with binary cross-entropy. The task-specific heads are joined by the multi-task objective $L_{\text{total}} = L_{\text{CE}} + \lambda_1 L_{\text{BCE}} + \lambda_2 L_{\text{rep}}$, where the non-repeat term $L_{\text{rep}}(\hat{Y}) = L - |\mathrm{unique}(\hat{Y})|$ is designed to push the decoder toward diverse concept sequences. This penalty is what carries the paper's claim of reducing redundant recommendations within a generated path.

What would settle it

Retrain the model on ASSIST09 with the non-repeat term $L_{\mathrm{rep}}$ removed but all other settings unchanged; if accuracy, F1, and duplicate rates stay the same, the penalty is not what drives the reported gains. A companion check is to compute the gradient of $L_{\mathrm{rep}}$ under the actual decoding rule and see whether it is nonzero at any training step.

Watch

Extended reading notes

Core claim

The central claim is that sharing one LSTM representation between next-item path prediction and a binary knowledge-tracing head, while adding a non-repeat penalty, produces better learning paths than separately trained sequence models. The model redefines learning path recommendation as the conditional probability $P_\theta(Y|X)=\prod_{t=1}^{m} P(y_t|y_{<t},X)$, computes a shared hidden sequence $H=\mathrm{LSTM}_\theta(X)$, and then uses task-specific LSTM heads to output the recommended path and a success/failure probability. The total loss is $L_{\text{total}} = L_{\text{CE}} + \lambda_1 L_{\text{BCE}} + \lambda_2 L_{\text{rep}}$, where $L_{\text{rep}}(\hat{Y}) = L - |\mathrm{unique}(\hat{Y})|$ penalizes repeated items. Reported results on ASSIST09 show the proposed method outperforming all six baselines across all evaluation metrics.

Load-bearing premise

The central premise is that the non-repeat penalty, which counts distinct items in a discrete predicted path, can be optimized by gradient descent; the paper gives no continuous approximation, so if that term cannot backpropagate, the claimed diversity benefit is unsupported.

Editorial extensions

If this is right

A multi-task objective with a shared LSTM can improve next-item path prediction over single-task RNN, LSTM, Seq2Seq, and attention baselines on a standard tutoring dataset.
The non-repeat penalty is intended to cut redundant recommendations, which, if it works, makes generated learning paths more usable in practice.
The Seq2Seq reframing provides a simple training objective for path generation that avoids reinforcement-learning machinery.
Longer target paths reduce accuracy for every model on ASSIST09, so the reported results suggest short paths of length 3 to 5 are the practical operating range.
The reported scores (accuracy 0.3489, F1 0.3241) offer a concrete comparison point for future learning path recommenders evaluated on ASSIST09.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

An implication the paper leaves implicit is that the multi-task gain may come largely from regularization: replacing the knowledge-tracing head with any unrelated binary auxiliary task and keeping the same shared LSTM could preserve most of the improvement, a claim that is directly testable.
A testable extension is to replace the discrete non-repeat penalty with a differentiable soft-uniqueness surrogate or to enforce no-repetition during decoding; the present discrete formulation may understate what a properly optimized diversity constraint could achieve.
Because the knowledge-tracing head receives only problem IDs, its correctness predictions lack the response and attempt features usually used for knowledge tracing; feeding correctness and attempt information into the shared encoder is a natural next step that could further improve path recommendation.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

This one's a near miss. The multi-task combination—sharing an LSTM between learning path recommendation and deep knowledge tracing, with a non-repeat penalty—is a legitimate extension to try, and the authors frame the recommendation task cleanly as Seq2Seq. Credit where due: they test on a standard benchmark, compare against six sensible baselines, and report a path-length breakdown. The high-level description of what they tried is clear.

The soft spots are load-bearing. The non-repeat loss in Eq. (7) is a count of unique discrete tokens; it is piecewise constant with zero gradient almost everywhere, so Adam cannot minimize it as claimed. No continuous surrogate or decoding-time enforcement is given. The deep-knowledge-tracing head in Eq. (6) requires binary correctness labels, while the input sequence X contains only problem IDs, with no correctness or attempt features. The paper never states where those y(i) come from or how the DKT head aligns with the path decoder. These are not cosmetic gaps; they mean the two components credited for the gain either do not produce a usable training signal or rely on labels that are never defined.

The empirical evidence is also under-supported. There is no train/test split, no error bars, no significance tests, and no ablation isolating the multi-task and non-repeat contributions. The reported accuracy gain over the best baseline is about 0.043, which could be real, but as presented it cannot be attributed to anything specific. The citations look fine—relevant prior work is there, and self-citation is not a problem.

Bottom line: the idea is worth a serious referee because the flaws are fixable in revision, but the current version is not publishable. The authors need to clarify the DKT inputs and labels, replace the non-repeat loss with a differentiable diversity penalty, provide a proper data split, and report repeated runs with ablations. I would not cite this as it stands, but I would read a revised version carefully.

Referee Report

4 major / 6 minor

Summary. The paper proposes a multi-task LSTM model for learning path recommendation. It reformulates the task as sequence-to-sequence prediction over problem IDs, adds a deep knowledge tracing head and a non-repeat loss, and reports improvements over six RNN/Seq2Seq/attention baselines on ASSIST09. The evaluation is limited to a single table of point estimates without error bars, significance tests, or ablations.

Significance. The problem is relevant and the idea of sharing an LSTM encoder between path prediction and knowledge tracing is reasonable. However, the paper as written does not define a trainable objective: the DKT head lacks specified inputs and labels, and the non-repeat loss is non-differentiable. The empirical support is a single table without train/test split details, standard deviations, or significance testing. As a result, the central claim is not supported and the method cannot be reproduced from the text.

major comments (4)

[Section III-B, Eq. (6) and Section IV-A] The deep knowledge tracing head is not well-defined. The input is described as a sequence of problem IDs only (X = {p1,...,pn}), with no correctness, attempt count, or response time, yet deep knowledge tracing (ref. [18]) requires (item, response) pairs to model a knowledge state. The paper never states where the ground-truth labels y(i) in Eq. (6) come from, nor how the DKT head is aligned with the path decoder. As written, the DKT head can at best learn per-problem difficulty, which is not knowledge tracing. This undermines the claimed multi-task gain.
[Section III-B, Eq. (7)] The non-repeat loss Lrep(Y_hat) = L - |unique(Y_hat)| is evaluated on discrete argmax outputs, making it piecewise constant with zero gradient almost everywhere. It cannot be minimized by the Adam optimizer as claimed, and the paper provides no continuous surrogate or decoding-time enforcement. This is load-bearing because the paper attributes part of the performance improvement to this regularizer.
[Section IV-D, Table II] The central empirical claim rests on a single table of point estimates. There is no description of a train/test split, no standard deviations, no statistical significance test, and no ablation isolating the multi-task head or the non-repeat loss. The word 'significantly' is therefore unsupported. Baseline implementations, hyperparameters, and data preprocessing are also unspecified, making the comparison irreproducible.
[Section IV-C, Eqs. (9)-(12)] The evaluation metrics are defined for binary classification, but the task is multi-class sequence prediction (predicting a path of problem IDs). The paper does not define how true positives, false positives, true negatives, and false negatives are computed over predicted paths, nor whether micro- or macro-averaging is used. Without this definition, the reported Accuracy, Precision, Recall, and F1 in Table II are ambiguous.

minor comments (6)

[Abstract and Section I] The abstract and introduction mention deep reinforcement learning, but the proposed method is a supervised Seq2Seq model. Please clarify the relation to RL or remove the RL framing if it is not used.
[Section I, Contribution 2] The contribution section claims evaluation with AUC, but Table II reports no AUC values. Either add AUC results or remove the claim.
[Figure 1] Figure 1 is referenced in Section IV-D but is not present in the provided manuscript text. If it appears in the PDF, please ensure it has labeled axes, a legend, and error bars or confidence intervals.
[Section III-B, Eq. (4)] Equation (4) uses g_phi(H) without defining the decoding mechanism. Please specify how the task-specific LSTM layers produce probability distributions over the target concept set, especially for the path recommendation head.
[Section III-B, Eq. (5)] The notation p_hat^{(i)}_{y^{(i)}} in Eq. (5) is confusing. Use standard cross-entropy notation such as -log p_hat(y_t | y_<t, X) and clarify the index ranges.
[Keywords] The keyword line contains a trailing semicolon and an inconsistent spacing in 'Multi Task Learning'. Please format keywords consistently.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported gains come from a standard supervised training and evaluation setup, not from an equation that reduces to its own input.

full rationale

The paper's central claim is an empirical performance comparison: a multi-task LSTM trained with loss components in Eqs. (5), (6), and (8) is evaluated on ASSIST09 and compared with six baselines in Table II. None of the equations defines a target quantity in terms of itself: Eq. (5) is a standard cross-entropy over next-item predictions, Eq. (6) is a standard binary cross-entropy for the auxiliary DKT-style head, and Eq. (7) is a regularizer over the predicted sequence. The proposal is not a theorem derived from assumptions that already contain the conclusion; it is a model trained under supervised learning and then tested on the same benchmark used for baselines. The references to multi-task learning [17] and deep knowledge tracing [18] are background methodology, not self-citations carrying load-bearing uniqueness arguments, and no step imports a result from the authors' own prior work. The implementation concerns raised by a close reading of Eqs. (6) and (7) — namely, that the input sequence contains only problem IDs while Eq. (6) requires binary success labels, and that the unique-count term in Eq. (7) is not obviously differentiable through argmax decoding — are internal-consistency and reproducibility issues, not circularity: they do not make any predicted quantity equal to a fitted input by construction. No fitted parameter is renamed as a prediction, no known result is repackaged under new coordinates, and no ansatz is smuggled in through a self-citation. The absence of ablations means the contribution of the multi-task head and the non-repeat penalty is not isolated, but that is an attribution gap, not a circular derivation. Accordingly, the circularity score is 0.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central method depends on the sequence-to-sequence modeling assumption, the ASSIST09 ground-truth assumption, the poorly specified DKT task, and the non-repeat loss as described. The main free parameters are the two loss weights, for which no values are given.

free parameters (2)

lambda1
Weight for the deep knowledge tracing loss L_BCE in Eq. (8). No value or tuning procedure is reported.
lambda2
Weight for the non-repeat loss L_rep in Eq. (8). No value or tuning procedure is reported.

assumptions (4)

domain assumption LSTM-based encoder-decoder architectures can accurately model the conditional probability P(Y|X) of a learning path given a student history (Eq. 1).
The whole approach rests on representing learning path generation as sequence prediction from problem IDs; this is asserted, not motivated with evidence.
domain assumption ASSIST09 problem-order sequences, sorted by order ID, provide valid ground-truth learning paths that the model should reproduce.
Section IV-A constructs X and Y directly from chronological problem logs, but actual study order may reflect system constraints rather than pedagogical intent.
ad hoc to paper The deep knowledge tracing task can be learned from an input sequence containing only problem IDs, with no correctness, attempt, or response-time features.
Section III-B defines DKT as binary success/failure classification, but Eq. (3) maps only X = problem IDs to hidden states; no correctness feature is introduced.
ad hoc to paper The non-repeat loss can be optimized with standard backpropagation despite involving |unique(Y_hat)|.
Eq. (7) uses a discrete set cardinality; no differentiable approximation is described.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing Learning Path Recommendation via Multi-task Learning." pith.science (2026). https://pith.science/paper/X4NLXUCK

@misc{pith2026250705295,
  author       = {Pith},
  title        = {Pith review of: Enhancing Learning Path Recommendation via Multi-task Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X4NLXUCK}},
  note         = {Machine review of arXiv:2507.05295}
}

read the original abstract

Personalized learning is a student-centered educational approach that adapts content, pace, and assessment to meet each learner's unique needs. As the key technique to implement the personalized learning, learning path recommendation sequentially recommends personalized learning items such as lectures and exercises. Advances in deep learning, particularly deep reinforcement learning, have made modeling such recommendations more practical and effective. This paper proposes a multi-task LSTM model that enhances learning path recommendation by leveraging shared information across tasks. The approach reframes learning path recommendation as a sequence-to-sequence (Seq2Seq) prediction problem, generating personalized learning paths from a learner's historical interactions. The model uses a shared LSTM layer to capture common features for both learning path recommendation and deep knowledge tracing, along with task-specific LSTM layers for each objective. To avoid redundant recommendations, a non-repeat loss penalizes repeated items within the recommended learning path. Experiments on the ASSIST09 dataset show that the proposed model significantly outperforms baseline methods for the learning path recommendation.

Figures

Figures reproduced from arXiv: 2507.05295 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 23 canonical work pages

[18]

Deep knowledge tracing,

C. Piech, J. Bassen, J. Huang, S. Ganguli, M. Sahami, L. J. Guibas, and J. Sohl-Dickstein, “Deep knowledge tracing,” Advances in neural information processing systems, vol. 28, 2015

work page 2015
[1]

Ai in education: A review of personalized learning and educational technology,

O. O. Ayeni, N. M. Al Hamad, O. N. Chisom, B. Osawaru, and O. E. Adewusi, “Ai in education: A review of personalized learning and educational technology,” GSC Advanced Research and Reviews, vol. 18, no. 2, 2024, pp. 261–271

work page 2024
[2]

A comprehensive analysis of personalized learning components,

A. Shemshack, Kinshuk, and J. M. Spector, “A comprehensive analysis of personalized learning components,” Journal of Computers in Educa- tion, vol. 8, no. 4, 2021, pp. 485–503

work page 2021
[3]

Components and strategies for personalized learning in higher education: A systematic,

I. Yuyun and D. Suherdi 1, “Components and strategies for personalized learning in higher education: A systematic,” in Proceedings of the 20th AsiaTEFL-68th TEFLIN-5th iNELTAL Conference (ASIATEFL 2022), vol. 749. Springer Nature, 2023, p. 271

work page 2022
[4]

Personalized learning systems: essential components and their impact on learning outcomes,

B. Wibawa, M. Wirasti et al., “Personalized learning systems: essential components and their impact on learning outcomes,” Journal of e- Learning and Knowledge Society, vol. 21, no. 1, 2025

work page 2025
[5]

Ai-supported health coaching model for patients with chronic diseases,

M. T. Sqalli and D. Al-Thani, “Ai-supported health coaching model for patients with chronic diseases,” in 2019 16th International Symposium on Wireless Communication Systems (ISWCS). IEEE, 2019, pp. 452– 456

work page 2019
[6]

Efficacy of an adaptive game- based math learning app to support personalized learning and improve early elementary school students’ learning,

H. J. Bang, L. Li, and K. Flynn, “Efficacy of an adaptive game- based math learning app to support personalized learning and improve early elementary school students’ learning,” Early Childhood Education Journal, vol. 51, no. 4, 2023, pp. 717–732

work page 2023
[7]

Fitness guide: A holistic approach for personalized health and wellness recom- mendation system,

A. Sathya, A. Vignesh, M. Akash, S. Gokulakrishnan et al., “Fitness guide: A holistic approach for personalized health and wellness recom- mendation system,” in 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). IEEE, 2024, pp. 01–06

work page 2024

Show all 30 references

[8]

Opti- mized deep learning framework for personalized nutritional recommen- dations across the menstrual cycle,

E. Logapriya, R. Surendran, D. Poornima, and L. Kartheesan, “Opti- mized deep learning framework for personalized nutritional recommen- dations across the menstrual cycle,” in 2024 International Conference on Sustainable Communication Networks and Application (ICSCNA). IEEE, 2...

2024
[9]

Full personalized learning path recommendation: A literature review,

N. Mansouri, M. Soui, and M. Abed, “Full personalized learning path recommendation: A literature review,” in International Conference on Advanced Intelligent Systems and Informatics. Springer, 2023, pp. 185–195

2023
[10]

A survey on learning path recommenda- tion,

F. Zhu, Y . Wen, and Q. Fu, “A survey on learning path recommenda- tion,” in CCF Conference on Computer Supported Cooperative Work and Social Computing. Springer, 2021, pp. 577–589

2021
[11]

A literature review of im- plemented recommendation techniques used in massive open online courses,

A. Khalid, K. Lundqvist, and A. Yates, “A literature review of im- plemented recommendation techniques used in massive open online courses,” Expert Systems with Applications, vol. 187, 2022, p. 115926

2022
[12]

Ld–lp generation of per- sonalized learning path based on learning diagnosis,

L. Meng, W. Zhang, Y . Chu, and M. Zhang, “Ld–lp generation of per- sonalized learning path based on learning diagnosis,” IEEE Transactions on Learning Technologies, vol. 14, no. 1, 2021, pp. 122–128

2021
[13]

Deep reinforcement learning: An overview,

Y . Li, “Deep reinforcement learning: An overview,” arXiv preprint arXiv:1701.07274, 2017

2017 arXiv
[14]

Set-to-sequence ranking-based concept- aware learning path recommendation,

X. Chen, J. Shen, W. Xia, J. Jin, Y . Song, W. Zhang, W. Liu, M. Zhu, R. Tang, K. Dong et al., “Set-to-sequence ranking-based concept- aware learning path recommendation,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 37, no. 4, 2023, pp. 5027– 5035

2023
[15]

Item- difficulty-aware learning path recommendation: From a real walking perspective,

H. Zhang, S. Shen, B. Xu, Z. Huang, J. Wu, J. Sha, and S. Wang, “Item- difficulty-aware learning path recommendation: From a real walking perspective,” in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024, pp. 4167–4178

2024
[16]

Privileged knowledge state distillation for reinforcement learning-based educational path recom- mendation,

Q. Li, W. Xia, L. Yin, J. Jin, and Y . Yu, “Privileged knowledge state distillation for reinforcement learning-based educational path recom- mendation,” in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024, pp. 1621–1630

2024
[17]

An overview of multi-task learning in deep neural networks,

S. Ruder, “An overview of multi-task learning in deep neural networks,” arXiv preprint arXiv:1706.05098, 2017

2017 arXiv
[19]

Recent advances in recurrent neural networks,

H. Salehinejad, S. Sankar, J. Barfett, E. Colak, and S. Valaee, “Recent advances in recurrent neural networks,” arXiv preprint arXiv:1801.01078, 2017

2017 arXiv
[20]

Attention, please! a survey of neural attention models in deep learning,

A. de Santana Correia and E. L. Colombini, “Attention, please! a survey of neural attention models in deep learning,” Artificial Intelligence Review, vol. 55, no. 8, 2022, pp. 6037–6124

2022
[21]

A survey on multi-task learning,

Y . Zhang and Q. Yang, “A survey on multi-task learning,” IEEE transactions on knowledge and data engineering, vol. 34, no. 12, 2021, pp. 5586–5609

2021
[22]

New types of deep neural network learning for speech recognition and related applications: An overview,

L. Deng, G. Hinton, and B. Kingsbury, “New types of deep neural network learning for speech recognition and related applications: An overview,” in 2013 IEEE international conference on acoustics, speech and signal processing. IEEE, 2013, pp. 8599–8603

2013
[23]

Massively multitask networks for drug discovery,

B. Ramsundar, S. Kearnes, P. Riley, D. Webster, D. Konerding, and V . Pande, “Massively multitask networks for drug discovery,” arXiv preprint arXiv:1502.02072, 2015

2015 arXiv
[24]

Fast r-cnn,

R. Girshick, “Fast r-cnn,” in Proceedings of the IEEE international conference on computer vision, 2015, pp. 1440–1448

2015
[25]

A unified architecture for natural language processing: Deep neural networks with multitask learning,

R. Collobert and J. Weston, “A unified architecture for natural language processing: Deep neural networks with multitask learning,” in Proceed- ings of the 25th international conference on Machine learning, 2008, pp. 160–167

2008
[26]

Addressing the assessment challenge with an online system that tutors as it assesses,

M. Feng, N. Heffernan, and K. Koedinger, “Addressing the assessment challenge with an online system that tutors as it assesses,” User modeling and user-adapted interaction, vol. 19, 2009, pp. 243–266

2009
[27]

Doubly constrained offline reinforcement learning for learning path recommendation,

Y . Yun, H. Dai, R. An, Y . Zhang, and X. Shang, “Doubly constrained offline reinforcement learning for learning path recommendation,” Knowledge-Based Systems, vol. 284, 2024, p. 111242

2024
[28]

On learning path planning algorithm based on collaborative analysis of learning behavior,

Z. Shou, X. Lu, Z. Wu, H. Yuan, H. Zhang, and J. Lai, “On learning path planning algorithm based on collaborative analysis of learning behavior,” IEEE Access, vol. 8, 2020, pp. 119 863–119 879

2020
[29]

Educational personalized learning path planning with large language models,

C. Ng and Y . Fung, “Educational personalized learning path planning with large language models,” arXiv preprint arXiv:2407.11773, 2024

2024 arXiv
[30]

Learn- ing structure and knowledge aware representation with large language models for concept recommendation,

Q. Li, W. Xia, K. Du, Q. Zhang, W. Zhang, R. Tang, and Y . Yu, “Learn- ing structure and knowledge aware representation with large language models for concept recommendation,” arXiv preprint arXiv:2405.12442, 2024

2024 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [18]

Deep knowledge tracing,

C. Piech, J. Bassen, J. Huang, S. Ganguli, M. Sahami, L. J. Guibas, and J. Sohl-Dickstein, “Deep knowledge tracing,” Advances in neural information processing systems, vol. 28, 2015

work page 2015

[2] [1]

Ai in education: A review of personalized learning and educational technology,

O. O. Ayeni, N. M. Al Hamad, O. N. Chisom, B. Osawaru, and O. E. Adewusi, “Ai in education: A review of personalized learning and educational technology,” GSC Advanced Research and Reviews, vol. 18, no. 2, 2024, pp. 261–271

work page 2024

[3] [2]

A comprehensive analysis of personalized learning components,

A. Shemshack, Kinshuk, and J. M. Spector, “A comprehensive analysis of personalized learning components,” Journal of Computers in Educa- tion, vol. 8, no. 4, 2021, pp. 485–503

work page 2021

[4] [3]

Components and strategies for personalized learning in higher education: A systematic,

I. Yuyun and D. Suherdi 1, “Components and strategies for personalized learning in higher education: A systematic,” in Proceedings of the 20th AsiaTEFL-68th TEFLIN-5th iNELTAL Conference (ASIATEFL 2022), vol. 749. Springer Nature, 2023, p. 271

work page 2022

[5] [4]

Personalized learning systems: essential components and their impact on learning outcomes,

B. Wibawa, M. Wirasti et al., “Personalized learning systems: essential components and their impact on learning outcomes,” Journal of e- Learning and Knowledge Society, vol. 21, no. 1, 2025

work page 2025

[6] [5]

Ai-supported health coaching model for patients with chronic diseases,

M. T. Sqalli and D. Al-Thani, “Ai-supported health coaching model for patients with chronic diseases,” in 2019 16th International Symposium on Wireless Communication Systems (ISWCS). IEEE, 2019, pp. 452– 456

work page 2019

[7] [6]

Efficacy of an adaptive game- based math learning app to support personalized learning and improve early elementary school students’ learning,

H. J. Bang, L. Li, and K. Flynn, “Efficacy of an adaptive game- based math learning app to support personalized learning and improve early elementary school students’ learning,” Early Childhood Education Journal, vol. 51, no. 4, 2023, pp. 717–732

work page 2023

[8] [7]

Fitness guide: A holistic approach for personalized health and wellness recom- mendation system,

A. Sathya, A. Vignesh, M. Akash, S. Gokulakrishnan et al., “Fitness guide: A holistic approach for personalized health and wellness recom- mendation system,” in 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). IEEE, 2024, pp. 01–06

work page 2024