Pith. sign in

REVIEW 4 major objections 6 minor 58 references

Optimizing Student Ability Assessment: A Hierarchy Constraint-Aware Cognitive Diagnosis Framework for Educational Contexts

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read This paper claims that constraining cognitive diagnosis by student hierarchy levels improves accuracy and interpretability, with HCD-NCDM outperforming NCDM on three PISA datasets.

desk verdict Plausible framework with consistently positive results, but the headline claim is currently undercut by a likely leakage route in the hierarchy labels. read the letter →

arxiv 2412.04488 v1 pith:OQ7K72EH submitted 2024-11-21 cs.CY cs.AI

Xinjie Sun , Qi Liu , Kai Zhang , Shuanghong Shen , Fei Wang , Yan Zhuang , Zheng Zhang , Weiyin Gong
show 3 more authors
Shijin Wang Lina Yang Xingying Huo
This is my paper · ORCID
classification cs.CYcs.AI
keywords cognitivediagnosishierarchyconstraintstudentabilityassessmentinterpretabilityattentionmechanismPISAdatasetknowledgestatemodelingeducationaldatamining
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

Cognitive diagnosis models estimate what knowledge concepts a student has mastered from their response records. The paper argues that these estimates are incomplete because they treat each student in isolation and ignore the student's relative position inside the cohort. To fix this, the paper proposes HCD, a framework that first assigns each student a hierarchy level from the mean and standard deviation of all students' average scores, then refines that level into a knowledge-proficiency vector through two attention modules, and finally fuses it with the student's personalized embedding. Plugging this constraint into IRT, MIRT, DINA and NCDM improves response prediction, with the largest reported jumps for NCDM: AUC rises from about 78.91, 79.21 and 80.61 to 81.16, 82.17 and 83.66 on PISA-Science, PISA-Read and PISA-Math. If the effect is real, any cognitive diagnosis model can be given a group-relative anchor that makes its ability estimates smoother and more consistent with the prior distribution of student performance.

What carries the argument

The load-bearing object is the hierarchy-constrained knowledge proficiency vector $\theta_{hierarchy}$, produced by a hierarchy mapping layer and two named attention modules: the intra-hierarchy convolution-enhanced attention layer (CEA), which applies convolution and attention over students at the same level to expose fine-grained differences in identical knowledge concepts, and the inter-hierarchy random-sampling attention layer (RSA), which samples one student's features from other levels and attends to them to capture cross-level differences. The final student state is a convex combination $\theta = \alpha_{pw}\theta_{hierarchy} + (1 - \alpha_{pw})\theta_{person}$, where $\alpha_{pw}$ is a learned per-student weight and $\theta_{person}$ is the personalized proficiency vector. This machinery is what lets a base cognitive diagnosis model, such as IRT, MIRT, DINA or NCDM, be constrained without changing its diagnostic output function.

What would settle it

Restrict hierarchy construction to the 70% training portion: compute each student's average score and the interval boundaries from training records only, then evaluate on the untouched 20% test split with the same HCD-NCDM configuration and compare AUC to the current reported numbers. If the HCD advantage over NCDM shrinks to near zero while NCDM's AUC stays constant, the hierarchy constraint's reported gains would be attributable to information leakage rather than to the constraint itself.

Watch

Extended reading notes

Core claim

The paper's central claim is that student knowledge state should be modeled as the combination of an absolute, personalized proficiency vector and a relative, hierarchy-constrained proficiency vector, and that doing so yields both better prediction and better interpretability. The hierarchy constraint is derived from a student's average score relative to the cohort: the mean and standard deviation of all students' average scores define score intervals, and each student receives a hierarchy label from those intervals. The label is embedded into a knowledge-concept vector, processed by an intra-level convolution-enhanced attention layer that models differences among students at the same level, and by an inter-hierarchy random-sampling attention layer that captures differences across levels. The two representations are fused with a per-student learned weight, and the resulting state is fed through the cognitive diagnosis function of any base model. The paper reports that HCD-NCDM's learned knowledge states are smoother and align more closely with the prior statistics of student scores than those of NCDM, and it presents this as evidence that hierarchy constraints make assessment results reasonable and fair.

Load-bearing premise

The load-bearing premise is that the hierarchy level assigned to each student is computed from the average score of all students, and the paper does not state whether those averages and the interval boundaries are computed before or after the 70/10/20 split; if the hierarchy uses test-set responses, the reported accuracy gains could partly come from the model being told a coarse summary of the very labels it is asked to predict.

Editorial extensions

If this is right

  • Any existing cognitive diagnosis model can be augmented with HCD's mapping and attention layers and retrained, yielding higher response prediction accuracy and knowledge states that track the group's distribution.
  • The hierarchy constraint dampens extreme ability estimates, so a student's high performance on one exam does not inflate their diagnosed mastery without regard to their overall record.
  • The ablation results indicate that the intra-level convolution-enhanced attention layer is the main driver of the predictive gain, meaning fine-grained modeling within a level matters more than cross-level modeling.
  • The per-student learned weight $\alpha_{pw}$ lets the model adapt how strongly each student's estimate is anchored to their group, which supports a range of behaviors from personalized to group-driven diagnosis.
  • On the smaller PISA-Read dataset the constraint still improves performance, suggesting the approach helps when interaction records are sparse.

Reading between the lines

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

  • Because the hierarchy labels are computed from average scores over the full dataset before the train/test split is described, part of the reported AUC gain could stem from the model receiving a coarse summary of the test labels it is asked to predict; re-running the pipeline with hierarchy construction restricted to the training split would settle this.
  • The hierarchy mapping uses only a scalar average score per student, so a natural extension is to build levels from per-knowledge-concept prior statistics, which could produce finer-grained constraints and possibly larger gains.
  • The inter-level random-sampling attention is a scalability choice; a deterministic or variance-reduced aggregation across all students in other levels would show how much of the benefit of RSA comes from the sampling rather than from the cross-level signal.
  • The DOA results show DINA has the highest interpretability but lower predictive accuracy, so HCD can be read as a recipe for pushing neural models toward DINA-like interpretability while keeping their predictive strength.
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.

Referee Report

4 major / 6 minor

Summary. The paper proposes HCD, a hierarchy constraint-aware cognitive diagnosis framework. It computes student hierarchy levels from average scores, learns intra- and inter-hierarchy attention representations, and combines these with personalized knowledge proficiency. The authors instantiate HCD with IRT, MIRT, DINA, and NCDM and report improved AUC/ACC/RMSE on three PISA 2015 subsets, plus analyses of knowledge-state distributions, DOA interpretability, and case studies. The central claim is that hierarchy constraints improve both predictive accuracy and interpretability.

Significance. If the reported gains are valid, HCD would be a general, plug-in enhancement for existing cognitive diagnosis models, with a concrete interpretability benefit. The paper's strengths include public PISA data, four base models, consistent gains across three datasets, ablations of the two attention modules, and a promise of released code and data. The main significance hinges on whether the hierarchy construction is leakage-free; the current manuscript does not yet establish that.

major comments (4)
  1. [Sec. 4.2.1 and Sec. 5.1-5.2] The hierarchy labels are constructed from full-data average scores before any train/test split is described. Equation (5) computes the mean and standard deviation over "all students" and Equation (6) maps each student's average score xi to a level, with no statement that this mapping is recomputed inside each training fold using only training responses. Because the average score is a monotone aggregate of the binary response labels that the model is trained to predict, a hierarchy embedding built from full-data averages can directly leak test-set information into the training features. This could plausibly explain the Table 3 gains (e.g., +2.25 to +3.05 AUC for HCD-NCDM). Please clarify the exact protocol, and if hierarchy construction uses test data, re-run the experiments with a train-only hierarchy mapping and report the resulting Table 3 and Figure 3.
  2. [Sec. 4.2.1] The score-interval boundaries B = {b0,...,bg} in Equation (6) are never defined. The paper states only that the framework is based on "score intervals" and later fixes g = 6 in Section 5.6. The choice of boundaries is a free parameter and, if the boundaries are estimated from the full dataset, it is subject to the same leakage concern as Comment 1. Specify the algorithm for setting bj (e.g., mean +/- multiples of sigma, quantiles) and state whether the boundaries are fixed a priori or estimated from the training portion of each fold.
  3. [Sec. 5.1 and Sec. 5.2] The evaluation protocol is inconsistent: Section 5.1 says the datasets were "randomly divided" into 70% training, 10% validation, and 20% testing, while Section 5.2 says a 5-fold cross-validation was used and "the final results represent the average of the 5 folds." These cannot both describe the same experiments. Clarify which protocol generated Table 3 and describe how the hierarchy mapping and hyperparameter selection interact with the folds.
  4. [Sec. 5.5] The interpretability claim in RQ2 that HCD-NCDM knowledge states "exhibit a higher degree of fit with the prior statistical distribution" is weakened by the fact that the prior is the same full-data average-score distribution used to construct the hierarchy labels. If the hierarchy mapping leaks test responses, then the closer alignment of the learned knowledge states with that prior is partly by construction. Report this analysis for a train-only hierarchy construction to demonstrate that the smoothing is not an artifact of leakage.
minor comments (6)
  1. [Sec. 5.6 and Fig. 6 caption] The text contains corrupted encoding fragments (e.g., "/uni00000033/uni0000002c/..." in Section 5.6 and in the Figure 6 caption); please repair the PDF/LaTeX encoding.
  2. [Sec. 5.5] The sentence "this aligns with the general of ability development in human educational contexts" is incomplete; insert the missing noun (e.g., "general pattern" or "general law").
  3. [Sec. 4.4] Equation (20) and the surrounding text say the loss is computed on "each response logs record in the test data"; this should presumably read "training data" or "training logs," since the test set should not be used for optimization.
  4. [Sec. 4.2.3] The symbol Kh is used both for a convolution kernel in Eq. (9) and for the key matrix in Eq. (10); use distinct notation (e.g., Kc for the convolution kernel) to avoid ambiguity.
  5. [Eq. (7)] The one-hot encoding O in Eq. (7) is defined as "1 if Eq. (6)", but Equation (6) defines the mapping f(xi)=l_j; make explicit that Oij=1 exactly when f(xi)=hj.
  6. [Sec. 3.1] The definition of the student test record L as triples (s,e,h) assumes the hierarchy h is known as input, but the hierarchy is later constructed from average scores in Section 4.2.1; clarify whether h is observed or computed by the hierarchy mapping layer.

Circularity Check

2 steps flagged · score 6.0 of 10

Hierarchy labels are built from full-data average scores before the 70/10/20 split, so the reported AUC gains and the RQ2 'fit to prior' evidence can reflect target leakage and by-construction alignment rather than an independent hierarchy mechanism.

  1. fitted input called prediction [Section 4.2.1, Eqs. (5)-(6); Section 5.1; Section 5.2]
    "Let X = {x1, x2, ..., xn} represent the average scores of all students, where xi denotes the average score of the i-th student. ... f(xi) = lj ⇔ bj ≤ xi < bj+1, ∀i ∈ [1, n], j ∈ [0, g]. (6) ... Finally, we randomly divided all datasets into 70% for training, 10% for validation, and 20% for testing to ensure the effectiveness of model evaluation."

    A student's average score xi is a monotone aggregate of the binary responses r that the model is trained to predict. Eq. (6) turns that average into a hierarchy label before any split is described, and the paper never states that Eqs. (5)-(6) are recomputed inside each 5-fold cross-validation fold using only training data. Through Eqs. (7)-(8), the hierarchy one-hot enters the predictor, so the model input can carry a coarse summary of the very labels it is asked to predict. The Table 3 AUC deltas for HCD-NCDM (+2.25, +2.96, +3.05) are small enough to be explained by this leakage, making the 'prediction' partially forced by construction rather than by the hierarchy-constraint mechanism.

  2. self definitional [Section 5.5 (RQ2), with Eq. (6)]
    "after incorporating hierarchy constraint perception, students' ability performances become smoother and exhibit a higher degree of fit with the prior statistical distribution. ... in the PISA-Math dataset, students with an ability level of 0.50 constitute about 24.0% of the prior distribution, while the NCDM model only learns 13.5%; conversely, the HCD-NCDM model accounts for 28.0% of students at this level, which is closer to the prior distribution."

    The 'prior distribution' used in RQ2 is the average-score distribution whose mean and standard deviation define the score intervals in Eq. (5) and whose interval labels are assigned in Eq. (6). Those labels are the direct inputs to HCD's hierarchy embedding (Eqs. (7)-(8)). Showing that HCD's learned states cluster near the same prior is therefore partly a consequence of the construction, not an independent confirmation of interpretability. The comparison retains some content because NCDM's states are learned freely, but the claimed 'higher degree of fit' is not an external check on the hierarchy constraint.

full rationale

The central predictive claim is tested against standard baselines and a nominal held-out split, so it is not fully circular. However, the paper defines hierarchy labels from 'the average scores of all students' before describing the 70/10/20 split, and it never states that the mean, standard deviation, or interval boundaries are recomputed on training data only. Because a student's average score is an aggregate of the very responses the model predicts, the hierarchy embedding can leak test-label information into the predictor; the Table 3 improvements may then reflect that leakage rather than a genuine hierarchy constraint. The RQ2 interpretability argument is similarly weakened: the 'prior statistical distribution' that HCD is said to fit is the same distribution used to construct the hierarchy input, so the fit is partly by construction. The six-level choice cites the authors' prior work (Sun et al., 2024c), but that choice is not load-bearing and does not by itself raise the score. Overall, the paper has substantial independent engineering, but the load-bearing evidence for the hierarchy constraint's benefit reduces in part to an unguarded full-data preprocessing step and a self-referential prior-alignment check.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The framework relies on several fitted or hand-chosen components, most importantly the hierarchy level count and interval boundaries, and on domain assumptions about the validity of average-score-based grouping. No new physical entities, forces, or conserved quantities are introduced; the hierarchy constraint is a modeling construct over existing student response data.

free parameters (4)
  • number of hierarchy levels g = 6 in experiments
    The paper never states g in the setup; Figure 4 and Section 5.6 reveal six levels. The choice is arbitrary and no sensitivity analysis is provided. It controls the granularity of the hierarchy constraint.
  • score interval boundaries B_j = not reported
    Eq. (5)-(6) define B via mean and standard deviation of average scores but do not specify how many intervals or where boundaries lie. This determines every student's hierarchy label and hence the entire model.
  • learned mixing weight alpha_pw = learned via sigma(E_person[s])
    The personalized versus hierarchical feature weighting is trained from data rather than fixed. This is a trainable parameter fitted on the same response logs as the target.
  • CEA and RSA hyperparameters = not reported
    Section 4.2.3-4.2.4 define the layers but omit kernel sizes, attention head count, and sampling strategy details. These choices affect the reported gains.
assumptions (5)
  • domain assumption The expert-annotated Q-matrix correctly represents the knowledge concepts required by each exercise.
    Section 3.1 takes Q as predetermined; errors in Q would distort all ability estimates.
  • domain assumption A student's normalized average score over all responses is a valid basis for assigning that student to an ability hierarchy.
    No independent evidence is given that average-score partitions map to true ability levels, yet this assumption is load-bearing for the hierarchy constraint.
  • domain assumption Students at the same hierarchy level are similar enough in overall ability that intra-level attention on shared knowledge concepts is meaningful.
    Section 4.2.3 motivates CEA with this similarity; if within-level variation is as large as between-level variation, the hierarchy constraint adds noise.
  • ad hoc to paper Randomly sampling one student from each other level yields a representative cross-level context for RSA.
    Section 4.2.4 says random sampling reduces computation, but no analysis shows that one sample per level suffices to capture cross-level differences.
  • domain assumption Random split of interaction records into 70/10/20 is a valid evaluation protocol for cognitive diagnosis.
    This assumes interaction-level splits generalize to the deployment scenario; if the goal is diagnosing new students, this evaluation may overstate performance.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimizing Student Ability Assessment: A Hierarchy Constraint-Aware Cognitive Diagnosis Framework for Educational Contexts." pith.science (2026). https://pith.science/paper/OQ7K72EH

@misc{pith2026241204488,
  author       = {Pith},
  title        = {Pith review of: Optimizing Student Ability Assessment: A Hierarchy Constraint-Aware Cognitive Diagnosis Framework for Educational Contexts},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OQ7K72EH}},
  note         = {Machine review of arXiv:2412.04488}
}
read the original abstract

Cognitive diagnosis (CD) aims to reveal students' proficiency in specific knowledge concepts. With the increasing adoption of intelligent education applications, accurately assessing students' knowledge mastery has become an urgent challenge. Although existing cognitive diagnosis frameworks enhance diagnostic accuracy by analyzing students' explicit response records, they primarily focus on individual knowledge state, failing to adequately reflect the relative ability performance of students within hierarchies. To address this, we propose the Hierarchy Constraint-Aware Cognitive Diagnosis Framework (HCD), designed to more accurately represent student ability performance within real educational contexts. Specifically, the framework introduces a hierarchy mapping layer to identify students' levels. It then employs a hierarchy convolution-enhanced attention layer for in-depth analysis of knowledge concepts performance among students at the same level, uncovering nuanced differences. A hierarchy inter-sampling attention layer captures performance differences across hierarchies, offering a comprehensive understanding of the relationships among students' knowledge state. Finally, through personalized diagnostic enhancement, the framework integrates hierarchy constraint perception features with existing models, improving the representation of both individual and group characteristics. This approach enables precise inference of students' knowledge state. Research shows that this framework not only reasonably constrains changes in students' knowledge states to align with real educational settings, but also supports the scientific rigor and fairness of educational assessments, thereby advancing the field of cognitive diagnosis.

Figures

Figures reproduced from arXiv: 2412.04488 by the authors.

Figure 1
Figure 1. A toy example of cognitive diagnosis, where CDM stands for cog￾nitive diagnosis model and GKP stands for group knowledge proficiency. understanding of their learning progress. As shown in [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of HCD Framework. the number of knowledge concepts). This process is typically based on manually designed cognitive behavior models, such as the item response functions used in IRT. To ensure that the predicted student ability values more ac￾curately reflect the students’ true capabilities, we consider rel￾ative constraints by comparing their performance with that of other students. For example, a student’s… view at source ↗
Figure 3
Figure 3. Comparison of kernel density plots for the mean score based on prior statistics, the mean knowledge state of students from the NCDM model, and the mean knowledge state of students from the HCD-NCDM model on the PISA-Read and PISA-Math datasets. • DINA (De La Torre, 2009): The DINA model assumes that each item involves multiple knowledge concepts and infers students’ mastery of these points by analyzing their respons… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Demonstrating knowledge concept Mastery for Students Across Six Levels in the PISA-Math Dataset: Random Selection of One Student from Each Level Several important findings emerge. First, overall, both the NCDM and the hierarchy constraint-enhanced HCD-NCDM mod￾els are …
Figure 6
Figure 6. Figure 6: Randomly selecting 30 students from the PISA-Science dataset and averaging their abilities across various knowledge concepts, we perform the fusion of the two ability characteristics according to Equation Eq.(17). e1 e2 e3 e4 e5 Exercise Student Response e1 e2 e3 e4 e5…
Figure 7
Figure 7. Figure 7: HCD-NCDM cognitive diagnosis case study: interrelationships among personalized ability θperson, hierarchical abilities θhierarchy, Overall Ability θ, and Item Difficulty. However, we also recognize that the current hierarchy con￾straint awareness primarily targets effe…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 55 canonical work pages

  1. [1]

    Multidimensional item response theory models

    Terry A Ackerman. Multidimensional item response theory models. Wiley StatsRef: Statistics Reference Online, 2014

    2014

  2. [2]

    Item response theory: What it is and how you can use the irt procedure to apply it

    Xinming An and Yiu-Fai Yung. Item response theory: What it is and how you can use the irt procedure to apply it. SAS Institute Inc, 10 0 (4): 0 364--2014, 2014

    work page 2014

  3. [3]

    Application of bloom’s taxonomy in categorization of cognitive process development in colleges

    Selina Banda, Florence Phiri, Jack Kaale, Agnes M Banda, DL Mpolomoka, R Chikopela, and C Mushibwe. Application of bloom’s taxonomy in categorization of cognitive process development in colleges. Journal of Education and Practice, 14 0 (4): 0 6--13, 2023

    work page 2023

  4. [4]

    A meta-analysis of blended learning and technology use in higher education: From the general to the applied

    Robert M Bernard, Eugene Borokhovski, Richard F Schmid, Rana M Tamim, and Philip C Abrami. A meta-analysis of blended learning and technology use in higher education: From the general to the applied. Journal of Computing in Higher Education, 26: 0 87--122, 2014

    work page 2014

  5. [5]

    Taxonomy of educational objectives: the classification of educational goals

    Benjamin S Bloom et al. Taxonomy of educational objectives: the classification of educational goals. susan fauer company, 1956

    work page 1956

  6. [6]

    mirt: A multidimensional item response theory package for the r environment

    R Philip Chalmers. mirt: A multidimensional item response theory package for the r environment. Journal of statistical Software, 48: 0 1--29, 2012

    work page 2012

  7. [7]

    Predicting matchups and preferences in context

    Shuo Chen and Thorsten Joachims. Predicting matchups and preferences in context. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 775--784, 2016

    work page 2016

  8. [8]

    Inferring the number of attributes for the exploratory dina model

    Yinghan Chen, Ying Liu, Steven Andrew Culpepper, and Yuguo Chen. Inferring the number of attributes for the exploratory dina model. Psychometrika, 86 0 (1): 0 30--64, 2021

    work page 2021

Show all 58 references
  1. [9]

    Tracking knowledge proficiency of students with educational priors

    Yuying Chen, Qi Liu, Zhenya Huang, Le Wu, Enhong Chen, Runze Wu, Yu Su, and Guoping Hu. Tracking knowledge proficiency of students with educational priors. In Proceedings of the 2017 ACM on Conference on Information and Knowledge Management, pages 989--998, 2017

    2017

  2. [10]

    Dirt: Deep learning enhanced item response theory for cognitive diagnosis

    Song Cheng, Qi Liu, Enhong Chen, Zai Huang, Zhenya Huang, Yiying Chen, Haiping Ma, and Guoping Hu. Dirt: Deep learning enhanced item response theory for cognitive diagnosis. In Proceedings of the 28th ACM international conference on information and knowledge management, pages ...

    2019

  3. [11]

    Dina model and parameter estimation: A didactic

    Jimmy De La Torre. Dina model and parameter estimation: A didactic. Journal of educational and behavioral statistics, 34 0 (1): 0 115--130, 2009

    2009

  4. [12]

    Towards a taxonomy of technological innovations in higher education

    Thandeka Dlamini and Aurona Gerber. Towards a taxonomy of technological innovations in higher education. pages 136--148, 2023

    2023

  5. [13]

    Intelligent tutoring systems: How computers achieve learning gains that rival human tutors

    Sidney K D'Mello and Art Graesser. Intelligent tutoring systems: How computers achieve learning gains that rival human tutors. In Handbook of educational psychology, pages 603--629. Routledge, 2023

    2023

  6. [14]

    Item parameter estimation for the mirt model: Bias and precision of confirmatory factor analysis—based models

    Holmes Finch. Item parameter estimation for the mirt model: Bias and precision of confirmatory factor analysis—based models. Applied Psychological Measurement, 34 0 (1): 0 10--26, 2010

    2010

  7. [15]

    Random-walk computation of similarities between nodes of a graph with application to collaborative recommendation

    Francois Fouss, Alain Pirotte, Jean-Michel Renders, and Marco Saerens. Random-walk computation of similarities between nodes of a graph with application to collaborative recommendation. IEEE Transactions on knowledge and data engineering, 19 0 (3): 0 355--369, 2007

    2007

  8. [16]

    Rcd: Relation map driven cognitive diagnosis for intelligent education systems

    Weibo Gao, Qi Liu, Zhenya Huang, Yu Yin, Haoyang Bi, Mu-Chun Wang, Jianhui Ma, Shijin Wang, and Yu Su. Rcd: Relation map driven cognitive diagnosis for intelligent education systems. In Proceedings of the 44th international ACM SIGIR conference on research and development in i...

    2021

  9. [17]

    Context-aware attentive knowledge tracing

    Aritra Ghosh, Neil Heffernan, and Andrew S Lan. Context-aware attentive knowledge tracing. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, pages 2330--2339, 2020

    2020

  10. [18]

    Using connectionist models to evaluate examinees’ response patterns to achievement tests

    Mark J Gierl, Ying Cui, and Steve Hunka. Using connectionist models to evaluate examinees’ response patterns to achievement tests. Journal of Modern Applied Statistical Methods, 7 0 (1): 0 19, 2008

    2008

  11. [19]

    Modeling physicians' utterances to explore diagnostic decision-making

    Xuan Guo, Rui Li, Qi Yu, and Anne R Haake. Modeling physicians' utterances to explore diagnostic decision-making. In IJCAI, pages 3700--3706, 2017

    2017

  12. [20]

    Parameters and models of item response theory (irt): A review of literature

    Abraham Gyamfi and Rosemary Acquaye. Parameters and models of item response theory (irt): A review of literature. Acta Educationis Generalis, 13 0 (3): 0 68--78, 2023

    2023

  13. [21]

    Enhancing computational thinking skills through artificial intelligence education at a steam high school

    Xiaodong Huang and Chengche Qiao. Enhancing computational thinking skills through artificial intelligence education at a steam high school. Science & Education, 33 0 (2): 0 383--403, 2024

    2024

  14. [22]

    Irt and mirt models for item parameter estimation with multidimensional multistage tests

    Paul A Jewsbury and Peter W van Rijn. Irt and mirt models for item parameter estimation with multidimensional multistage tests. Journal of Educational and Behavioral Statistics, 45 0 (4): 0 383--402, 2020

    2020

  15. [23]

    Augmented reality application in classroom: an immersive taxonomy

    R Kaviyaraj and M Uma. Augmented reality application in classroom: an immersive taxonomy. pages 1221--1226, 2022

    2022

  16. [24]

    A taxonomy of factors affecting attitudes towards educational technologies for use with technology acceptance models

    Andrew Kemp, Edward Palmer, and Peter Strelan. A taxonomy of factors affecting attitudes towards educational technologies for use with technology acceptance models. British Journal of Educational Technology, 50 0 (5): 0 2394--2413, 2019

    2019

  17. [25]

    Aligning open educational resources to new taxonomies: How ai technologies can help and in which scenarios

    Zhi Li, Zachary A Pardos, and Cheng Ren. Aligning open educational resources to new taxonomies: How ai technologies can help and in which scenarios. Computers & Education, 216: 0 105027, 2024

    2024

  18. [26]

    Towards a new generation of cognitive diagnosis

    Qi Liu. Towards a new generation of cognitive diagnosis. In IJCAI, pages 4961--4964, 2021

    2021

  19. [27]

    Fuzzy cognitive diagnosis for modelling examinee performance

    Qi Liu, Runze Wu, Enhong Chen, Guandong Xu, Yu Su, Zhigang Chen, and Guoping Hu. Fuzzy cognitive diagnosis for modelling examinee performance. ACM Transactions on Intelligent Systems and Technology (TIST), 9 0 (4): 0 1--26, 2018

    2018

  20. [28]

    Ability testing & talent identification

    David F Lohman and Megan Foley Nicpon. Ability testing & talent identification. In Identification, pages 287--335. Routledge, 2023

    2023

  21. [29]

    A theory of test scores

    Frederic Lord. A theory of test scores. Psychometric monographs, 1952

    1952

  22. [30]

    Applications of item response theory to practical testing problems

    Frederic M Lord. Applications of item response theory to practical testing problems. Routledge, 2012

    2012

  23. [31]

    A theory of human motivation

    AH Maslow. A theory of human motivation. Psychological Review google schola, 2: 0 21--28, 1943

    1943

  24. [32]

    A systematic review of analytic hierarchy process applications to solve transportation problems: from 2003 to 2022

    Sarbast Moslem, Mahyar Kamali Saraji, Abbas Mardani, Ahmad Alkharabsheh, Szabolcs Duleba, and Domokos Eszterg \'a r-Kiss. A systematic review of analytic hierarchy process applications to solve transportation problems: from 2003 to 2022. Ieee Access, 11: 0 11973--11990, 2023

    2003

  25. [33]

    Construction and evaluation of a web application for the educational process on normal distribution considering the science of data and machine learning

    Ricardo-Ad \'a n Salas-Rueda. Construction and evaluation of a web application for the educational process on normal distribution considering the science of data and machine learning. Research in Learning Technology, 27, 2019

    2019

  26. [34]

    Assessing student's dynamic knowledge state by exploring the question difficulty effect

    Shuanghong Shen, Zhenya Huang, Qi Liu, Yu Su, Shijin Wang, and Enhong Chen. Assessing student's dynamic knowledge state by exploring the question difficulty effect. In Proceedings of the 45th international ACM SIGIR conference on research and development in information retriev...

    2022

  27. [35]

    Affect and inference in bayesian knowledge tracing with a robot tutor

    Samuel Spaulding and Cynthia Breazeal. Affect and inference in bayesian knowledge tracing with a robot tutor. In Proceedings of the Tenth Annual ACM/IEEE international conference on human-robot interaction extended abstracts, pages 219--220, 2015

    2015

  28. [36]

    Using hierarchical irt models to create unidimensional measures from multidimensional data

    Brian D Stucky and Maria Orlando Edelen. Using hierarchical irt models to create unidimensional measures from multidimensional data. Handbook of item response theory modeling, pages 183--206, 2014

    2014

  29. [37]

    Bpcd: Bidirectional parallel enhanced cognitive diagnostic model integrating theory and experiment

    Xinjie Sun, Weiyin Gong, Dongping Yong, Guoying Sun, Nali Yang, and Hui Li. Bpcd: Bidirectional parallel enhanced cognitive diagnostic model integrating theory and experiment. Available at SSRN 4979153, 2024 a

    2024

  30. [38]

    Lgs-kt: Integrating logical and grammatical skills for effective programming knowledge tracing

    Xinjie Sun, Shuanghong Shen, Yan Zhuang, Yuxiang Guo, et al. Lgs-kt: Integrating logical and grammatical skills for effective programming knowledge tracing. Neural Networks, 2024 b

    2024

  31. [39]

    Target hierarchy-guided knowledge tracing: Fine-grained knowledge state modeling

    Xinjie Sun, Kai Zhang, Shuanghong Shen, Fei Wang, Yuxiang Guo, and Qi Liu. Target hierarchy-guided knowledge tracing: Fine-grained knowledge state modeling. Expert Systems with Applications, 251: 0 123898, 2024 c

    2024

  32. [40]

    Rule space: An approach for dealing with misconceptions based on item response theory

    Kikumi K Tatsuoka. Rule space: An approach for dealing with misconceptions based on item response theory. Journal of educational measurement, pages 345--354, 1983

    1983

  33. [41]

    Definition of student competencies and development of an educational plan to assess student mastery level

    Ronald E Terry, John N Harb, William C Hecker, and W Vincent Wilding. Definition of student competencies and development of an educational plan to assess student mastery level. International Journal of Engineering Education, 18 0 (2): 0 225--235, 2002

    2002

  34. [42]

    Multi-relational factorization models for student modeling in intelligent tutoring systems

    Nguyen Thai-Nghe and Lars Schmidt-Thieme. Multi-relational factorization models for student modeling in intelligent tutoring systems. In 2015 Seventh international conference on knowledge and systems engineering (KSE), pages 61--66. IEEE, 2015

    2015

  35. [43]

    Recommender system for predicting student performance

    Nguyen Thai-Nghe, Lucas Drumond, Artus Krohn-Grimberghe, and Lars Schmidt-Thieme. Recommender system for predicting student performance. Procedia Computer Science, 1 0 (2): 0 2811--2819, 2010

    2010

  36. [44]

    A survey of personalized adaptive learning system

    Bin Tian, Chao Wang, and Hanyu Hong. A survey of personalized adaptive learning system. In 2023 2nd International Conference on Artificial Intelligence and Computer Information Technology (AICIT), pages 1--6. IEEE, 2023

    2023

  37. [45]

    Collaborative filtering applied to educational data mining

    Andreas Toscher and Michael Jahrer. Collaborative filtering applied to educational data mining. KDD cup, 2010

    2010

  38. [46]

    Neural cognitive diagnosis for intelligent education systems

    Fei Wang, Qi Liu, Enhong Chen, Zhenya Huang, Yuying Chen, Yu Yin, Zai Huang, and Shijin Wang. Neural cognitive diagnosis for intelligent education systems. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 6153--6161, 2020

    2020

  39. [47]

    Neuralcd: a general framework for cognitive diagnosis

    Fei Wang, Qi Liu, Enhong Chen, Zhenya Huang, Yu Yin, Shijin Wang, and Yu Su. Neuralcd: a general framework for cognitive diagnosis. IEEE Transactions on Knowledge and Data Engineering, 35 0 (8): 0 8312--8327, 2022

    2022

  40. [48]

    A survey of models for cognitive diagnosis: New developments and future directions

    Fei Wang, Weibo Gao, Qi Liu, Jiatong Li, Guanhao Zhao, Zheng Zhang, Zhenya Huang, Mengxiao Zhu, Shijin Wang, Wei Tong, et al. A survey of models for cognitive diagnosis: New developments and future directions. arXiv preprint arXiv:2407.05458, 2024 a

    2024 arXiv

  41. [49]

    Self-supervised graph learning for long-tailed cognitive diagnosis

    Shanshan Wang, Zhen Zeng, Xun Yang, and Xingyi Zhang. Self-supervised graph learning for long-tailed cognitive diagnosis. In Proceedings of the AAAI conference on artificial intelligence, volume 37, pages 110--118, 2023

    2023

  42. [50]

    Boosting neural cognitive diagnosis with student’s affective state modeling

    Shanshan Wang, Zhen Zeng, Xun Yang, Ke Xu, and Xingyi Zhang. Boosting neural cognitive diagnosis with student’s affective state modeling. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 620--627, 2024 b

    2024

  43. [51]

    Echo: a layered model for the design of a context-aware learning experience

    Hadas Weinberger. Echo: a layered model for the design of a context-aware learning experience. In Handbook of Research on Web 2.0, 3.0, and X. 0: Technologies, Business, and Social Applications, pages 539--555. IGI Global, 2010

    2010

  44. [52]

    Contextualism as a world view for the reformation of meetings

    John Whiteside and Dennis Wixon. Contextualism as a world view for the reformation of meetings. In Proceedings of the 1988 ACM conference on Computer-supported cooperative work, pages 369--376, 1988

    1988

  45. [53]

    Variational bayes inference for the dina model

    Kazuhiro Yamaguchi and Kensuke Okada. Variational bayes inference for the dina model. Journal of Educational and Behavioral Statistics, 45 0 (5): 0 569--597, 2020

    2020

  46. [54]

    Gikt: a graph-based interaction model for knowledge tracing

    Yang Yang, Jian Shen, Yanru Qu, Yunfei Liu, Kerong Wang, Yaoming Zhu, Weinan Zhang, and Yong Yu. Gikt: a graph-based interaction model for knowledge tracing. In Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2020, Ghent, Belgium, Septembe...

    2020

  47. [55]

    Exploiting non-interactive exercises in cognitive diagnosis

    Fangzhou Yao, Qi Liu, Min Hou, Shiwei Tong, Zhenya Huang, Enhong Chen, Jing Sha, and Shijin Wang. Exploiting non-interactive exercises in cognitive diagnosis. Interaction, 100 0 (200): 0 300, 2023

    2023

  48. [56]

    Multi-factors aware dual-attentional knowledge tracing

    Moyu Zhang, Xinning Zhu, Chunhong Zhang, Yang Ji, Feng Pan, and Changchuan Yin. Multi-factors aware dual-attentional knowledge tracing. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management, pages 2588--2597, 2021

    2021

  49. [57]

    Understanding and improving fairness in cognitive diagnosis

    Zheng Zhang, Le Wu, Qi Liu, Jiayu Liu, Zhenya Huang, Yu Yin, Yan Zhuang, Weibo Gao, and Enhong Chen. Understanding and improving fairness in cognitive diagnosis. Science China Information Sciences, 67 0 (5): 0 152106, 2024

    2024

  50. [58]

    Modeling context-aware features for cognitive diagnosis in student learning

    Yuqiang Zhou, Qi Liu, Jinze Wu, Fei Wang, Zhenya Huang, Wei Tong, Hui Xiong, Enhong Chen, and Jianhui Ma. Modeling context-aware features for cognitive diagnosis in student learning. In Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, pages 2...

    2021

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.