Pith. sign in

REVIEW 5 major objections 5 minor 47 references

Social Influence-based Attentive Mavens Mining and Aggregative Representation Learning for Group Recommendation

T0 review · 5 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A group recommender that learns which members steer each decision and encodes the group as a sentence claims better accuracy than existing baselines.

desk verdict A plausible incremental model for group recommendation whose central BERT-based claim is under-specified to the point of being untestable, though the attention-maven idea has merit. read the letter →

arxiv 1909.01079 v1 pith:HPPJM7AJ submitted 2019-08-10 cs.SI cs.IR

Peipei Wang , Lin Li , Yi Yu , Guandong Xu This is my paper · ORCID
classification cs.SIcs.IR
keywords grouprecommendationsocialinfluenceattentionmechanismmavensminingBERTrepresentationneuralcollaborativefilteringprofilemodeling
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 group members do not contribute equally to a group decision, and that learning who matters per decision, rather than applying a fixed rule like averaging, produces better group recommendations. It grounds this weighting in social identity theory and two-step flow theory, then uses a pretrained BERT model to encode the group as a sentence vector that captures member interaction. The resulting method, SIAGR, is claimed to beat the attention-based AGREE and several static aggregation baselines on two real-world datasets in hit ratio and mean reciprocal rank. If right, this would mean dynamic, theoretically motivated preference aggregation can replace hand-set strategies in group recommenders.

What carries the argument

The load-bearing object is the final group embedding $g_l(t)=\sum_j \alpha(t,j)\,u_j + g'_l$, where $\alpha(t,j)$ is a softmax attention weight computed from user and item embeddings, and $g'_l$ is a BERT sentence vector obtained by treating the group as a sentence whose words stand for members. The attention term instantiates the maven-mining idea derived from social identity theory and two-step flow theory; the BERT term is meant to capture interaction among members that a plain sum or average would miss. Both vectors are fed through an NCF-style multilayer network to score candidate items.

What would settle it

A controlled run of SIAGR with the BERT branch replaced by mean-pooling over member embeddings, keeping the attention maven branch and the NCF head unchanged, on the same CAMRa2011 and Plancast splits: if HR and MRR do not drop, the BERT interaction vector is not carrying the claimed signal.

Watch

Extended reading notes

Core claim

The paper's central claim is that representing a group as a weighted combination of a maven embedding and a BERT sentence vector, optimized inside neural collaborative filtering, yields more accurate group recommendations than fixed aggregation or attention-only baselines. Specifically, SIAGR computes per-member attention weights that depend on both the member's embedding and the candidate item, identifies the high-weight member as the maven, and adds a BERT-generated group vector to that weighted sum. The reported experiments on CAMRa2011 and Plancast show SIAGR outperforming AGREE, COM, PIT, NCF+AVG, and NCF+LM on HR and MRR.

Load-bearing premise

The load-bearing premise is that BERT, a language model pretrained on sentences, can be fed a set of user IDs as if it were a sentence and return a group vector that captures interactions among members; the paper does not specify how member IDs become tokens or whether BERT is fine-tuned.

Editorial extensions

If this is right

  • Learned per-item attention can replace fixed rules such as average or least misery in group recommenders, since SIAGR outperforms both NCF+AVG and NCF+LM.
  • Encoding the group as a sentence lets a pretrained contextual model contribute a group-level vector, and the aggregation of that vector with the maven embedding outperforms either component alone.
  • On both a small household movie dataset and a large event dataset, the proposed SIAGR reports higher HR and MRR than AGREE, the closest attention-based competitor.
  • The learned attention weights offer a per-decision picture of which member acts as maven, rather than one fixed influence score for all items.

Reading between the lines

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

  • If the BERT group vector really captures member interaction, the same sentence-style encoding could transfer to ephemeral groups, where a group has no history and the vector would be built from member IDs alone; the paper does not test cold-start groups and lists this as a limitation.
  • The attention weights could serve as post-hoc explanations of group decisions, pointing to the member who most shaped a recommendation; the paper visualizes such weights but does not evaluate them as explanations.
  • A separate question the paper does not settle is whether the gains come from the sociological weighting or simply from the extra parameters BERT adds; a capacity-matched baseline would separate the two.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 5 minor

Summary. The paper proposes SIAGR, a group recommendation method that combines an attention-based mavens mining mechanism, justified by social identity theory and two-step flow theory, with a BERT-based group embedding, and then feeds the aggregated representation into a neural collaborative filtering (NCF) framework. The authors claim that this combination yields better Hit Ratio and Mean Reciprocal Rank than AGREE, COM, PIT, NCF+AVG, and NCF+LM on CAMRa2011 and Plancast. The manuscript also includes visualizations of learned attention weights and of the relative contributions of the BERT-based and mavens-based components.

Significance. If the central claims were fully supported, the main contribution would be a plausible and novel integration of sociological theory with deep learning for group preference aggregation, specifically the idea of treating a group as a sentence and using BERT to capture member interactions. The paper makes a useful conceptual move by grounding mavens mining in explicit sociological theories rather than ad hoc intuitions, and the NCF-based framework is a reasonable backbone for the task. However, the empirical support is presently only qualitative (figures without numerical values), the BERT component is underspecified to the point of being irreproducible, and part of the interpretation of the attention weights is circular. The paper therefore illustrates an interesting research direction but, in its current form, does not yet substantiate the advertised performance improvement.

major comments (5)
  1. [§IV.B.3, Figs. 9–10] The central claim that "our method is better than AGREE in the two datasets" is supported only by figures. No numerical HR or MRR values, no standard deviations, and no significance tests are reported. Please provide tables of exact metrics for all methods on both datasets, along with error bars or significance tests, so that the reader can verify the magnitude and reliability of the improvement.
  2. [§III.C, Eq. (3)] The BERT-based group embedding is underspecified to the point of being untestable. The paper does not state how user IDs are mapped to BERT input tokens, whether BERT is frozen or fine-tuned on the group-item prediction objective, how variable-length groups are padded or truncated, which BERT variant is used, or which pooling operation yields the group vector. Without these details, Eq. (3)'s sum of the BERT group vector and the attention-based mavens embedding cannot be reproduced, and the claimed edge over AGREE cannot be attributed to BERT's contextual representation as opposed to the other trainable components.
  3. [§III.B, §IV.B.1, Figs. 3 and 6] The interpretation that learned attention weights "prove" the sociological claim that mavens matter is circular, because mavens are operationalized as the members with the highest learned attention weights. The visualization shows that the model assigns different weights, but it does not validate the theory or the existence of mavens in an external sense. Please rephrase this as a demonstration of the model's behavior, or validate the weights against an independent influence measure, user study, or ablation that changes the role of the identified mavens.
  4. [§IV.A.3 and reference [46]] The experiments omit the closest related work, reference [46] ("Social influence-based group representation learning for group recommendation," ICDE 2019), which the authors cite but do not compare against. Since this method also uses social influence and representation learning for group recommendation, the claimed state-of-the-art comparison is incomplete. Please include this baseline or explain why it cannot be compared.
  5. [§III.D, Eq. (7)] The pairwise loss in Eq. (7) is not clearly defined: the notation with repeated sums over O' and the squared terms suggests a regression-style pairwise objective, but the exact role of the negative sample s, the sampling strategy, and the relationship to the ranking metrics HR and MRR are unclear. Please rewrite the objective with explicit indices and state how negative instances are sampled during training.
minor comments (5)
  1. [Abstract and §I] There are several typos, including "To best our knowledge" in the contributions list and "our SIGAR" instead of "our SIAGR" in Section IV.B.3. A careful proofread is needed.
  2. [§IV.A.1] The dataset description says "The second dataset is CAMRa2011" when it is apparently the first dataset discussed; please reorder or renumber the dataset descriptions for clarity.
  3. [§III.B] The connection between social identity theory/two-step flow theory and the specific attention equation is only stated narratively. Please make the mapping explicit, e.g., which theoretical construct corresponds to the attention score, the softmax normalization, and the item-dependent conditioning.
  4. [§III.C] The phrase "What is worth mentioning, A word is not a group member and may have several words represent group member" is unclear and appears to be a fragment. Please rewrite this sentence to explain how, if at all, group members are tokenized for BERT.
  5. [§IV.B.2, Figs. 7–8] The ablation comparing SIAGR-G and SIAGR-M is presented as bar charts without numerical values or error bars, making it impossible to assess whether the aggregation gain is statistically meaningful. Including a small table with exact numbers would strengthen the RQ2 discussion.

Circularity Check

1 steps flagged · score 5.0 of 10

Maven-mining validation is self-definitional, but the central performance comparison is independent; BERT underspecification is a reproducibility issue, not circularity.

  1. self definitional [Section III.C, paragraph following Eq. (1) and Figure 3; also Section IV.B.1 (RQ1)]
    "With the attention mechanism, each member can contribute in group decision-making. A good example to illustrate this point as shown in Figure 3, users 1 and 12 have the largest attention weights or the largest influence as mavens in group 6, which are indicated by their darkest cells. More importantly, this approach corresponds to our theoretical support. Through our method, mavens of group can mining effectively and group members’ different influence can also reflect clearly."

    The paper operationalizes a maven as the group member with the largest learned attention weight. The attention weights are fitted parameters of the model, so localizing 'mavens' by taking the argmax of those weights is true by construction. Consequently, RQ1's demonstration that the learned weights identify mavens does not independently validate the two-step-flow claim that influential mavens drive group decisions; it only displays the fitted attention distribution. This interpretive step is circular, but it does not determine the empirical comparison with AGREE, COM, PIT, NCF+AVG, or NCF+LM, which is an external benchmark.

full rationale

The paper's strongest claim is empirical: 'our method is better than AGREE in the two datasets' (Section IV.B.3), and the results are benchmarked against five baselines. That comparison is self-contained and not derived from the paper's own assumptions by construction. The only load-bearing circular step is the maven-mining validation: since mavens are defined as the members with the largest attention weights, Figure 3 and RQ1 merely display the fitted attention distribution rather than providing independent evidence for the social-identity/two-step-flow rationale. This circularity weakens the interpretive claim that sociological theory is confirmed, but it does not force the empirical ranking. The BERT-based group embedding (Section III.C) is seriously underspecified - the paper does not state how user IDs are tokenized, whether BERT is frozen or fine-tuned, or which pooling produces the sentence vector - but underspecification is a reproducibility and correctness risk, not a circular reduction; no equation or citation makes the BERT output equal to its input by construction. No load-bearing self-citation chain appears: [45] is used only for dataset provenance and [11]/[44] are external baselines or tools. Overall score 5 reflects one self-definitional step in a supporting contribution while the central experimental claim remains independent.

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

The central claim rests on the validity of the BERT-as-group-encoder assumption, on the transferability of pretrained language representations to user ID sets, and on the adequacy of the evaluation. No code or data artifacts are shipped, so the upstream assumptions are not independently verified.

free parameters (4)
  • Attention network parameters (H_v, H_u, A, b) = learned from data
    Used in Eq. 1 to compute attention weights for mavens mining; fitted during training, not derived from theory.
  • NCF hidden and prediction layer weights (W_n, b_n, w) = learned from data
    Used in Eqs. 5-6 to predict group-item interactions; fitted during training.
  • BERT model parameters = pretrained or fine-tuned (unspecified)
    The paper uses BERT to generate group sentence vectors but does not specify whether frozen or fine-tuned; if fine-tuned, these are additional fitted parameters. The mapping of member IDs to tokens is also unspecified.
  • Embedding dimension, learning rate, number of layers, batch size = not reported
    Standard hyperparameters required to reproduce results; the paper does not report them.
assumptions (4)
  • domain assumption Social Identity Theory and Two-Step Flow Theory imply the existence of influential mavens in groups.
    Invoked in Section III.B to justify attention-based mavens mining; the theories are not formally modeled and no falsifiable prediction is derived from them.
  • ad hoc to paper A group can be represented as a sentence and its members as words for BERT encoding.
    Introduced in Section III.C; the analogy from natural language to group membership is not justified and the token mapping is unspecified.
  • domain assumption Pretrained BERT sentence embeddings encode meaningful interactions among group members.
    Relied on in Section III.C; no evidence is given that BERT representations transfer from text to user IDs.
  • standard math NCF is a suitable general framework for group recommendation.
    Borrowed from prior work [37] and used as the interaction learning framework in Section III.D.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Social Influence-based Attentive Mavens Mining and Aggregative Representation Learning for Group Recommendation." pith.science (2026). https://pith.science/paper/HPPJM7AJ

@misc{pith2026190901079,
  author       = {Pith},
  title        = {Pith review of: Social Influence-based Attentive Mavens Mining and Aggregative Representation Learning for Group Recommendation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HPPJM7AJ}},
  note         = {Machine review of arXiv:1909.01079}
}
read the original abstract

Frequent group activities of human beings have become an indispensable part in their daily life. Group recommendation can recommend satisfactory activities to group members in the recommender systems, and the key issue is how to aggregate preferences in different group members. Most existing group recommendation employed the predefined static aggregation strategies to aggregate the preferences of different group members, but these static strategies cannot simulate the dynamic group decision-making. Meanwhile, most of these methods depend on intuitions or assumptions to analyze the influence of group members and lack of convincing theoretical support. We argue that the influence of group members plays a particularly important role in group decision-making and it can better assist group profile modeling and perform more accurate group recommendation. To tackle the issue of preference aggregation for group recommendation, we propose a novel attentive aggregation representation learning method based on sociological theory for group recommendation, namely SIAGR (short for "Social Influence-based Attentive Group Recommendation"), which takes attention mechanisms and the popular method (BERT) as the aggregation representation for group profile modeling. Specifically, we analyze the influence of group members based on social identity theory and two-step flow theory and exploit an attentive mavens mining method. In addition, we develop a BERT-based representation method to learn the interaction of group members. Lastly, we complete the group recommendation under the neural collaborative filtering framework and verify the effectiveness of the proposed method by experimenting.

Figures

Figures reproduced from arXiv: 1909.01079 by the authors.

Figure 1
Figure 1. Illustration of the input data for group recommendation task, including user-item interactions, group-item interactions and user-user interactions We assume that there are a set of users U, a set of groups G and a set of items V in group recommendation. And there are three interaction among these data, namely user-item intera￾ction P, group-item interaction Q, user-user interaction R. We use P, Q, R represent user-i… view at source ↗
Figure 2
Figure 2. Illustration of attentive mavens mining method based on sociological theory [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Visualization for the sampled 10 groups w.r.t. attention weights, [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 46 canonical work pages

  1. [46]

    Social influence -based group representation learning for group recommendation

    Yin H, Wang Q, Zheng K, et al. Social influence -based group representation learning for group recommendation. In ICDE, 2019

    work page 2019

  2. [1]

    M. R. McLaughlin and J. L. Herlocker. A collaborative filtering algorithm and evaluation metric that accurately model the user experience. In SIGIR, pages 329–336, 2004

    work page 2004

  3. [2]

    B. M. Sarwar, G. Karypis, J. A. Konstan, and J. Riedl. I tem-based collaborative filtering recommendation algorithms. In WWW, pages 285–295, 2001

    work page 2001

  4. [3]

    Koren, R

    Y. Koren, R. M. Bell, and C. Volinsky. Matrix factorization techniques for recommender systems. IEEE Computer, 42(8):30–37, 2009

    work page 2009

  5. [4]

    Multivie w Spectral Clustering via Structured Low -Rank Matrix Factorization

    Y Wang, L Wu, X Lin, J Gao. Multivie w Spectral Clustering via Structured Low -Rank Matrix Factorization. IEEE Transactions on Neural Networks and Learning Systems 29 (10), 4833-4843, 2018

    work page 2018

  6. [5]

    Q. Yuan, G. Cong, and C.-Y. Lin. Com: a generative model for group recommendation. In SIGKDD, pages 163–172, 2014

    work page 2014

  7. [6]

    Berkovsky and J

    S. Berkovsky and J. Freyne. Group -based recipe recommendations: analysis of data aggregation strategies. In RecSys, pages 111 –118, 2010

    work page 2010

  8. [7]

    Group recommendations with rank aggregation and collaborative filtering

    Linas Baltrunas, Tadas Makcinskas, and Francesco Ricci. Group recommendations with rank aggregation and collaborative filtering. In RecSys, pages 119–126, 2010

    work page 2010

Show all 47 references
  1. [8]

    Group Recommendation: Semantics and Efficiency

    Sihem Amer-Yahia, Senjuti Basu Roy, Ashish Chawla, Gautam Das, and Cong Yu. Group Recommendation: Semantics and Efficiency. In VLDB, pages 754–765, 2009

    2009

  2. [9]

    State -of-the-Art in Group Recommendation and New Approaches for Automatic Identification of Groups

    Ludovico Boratto and Salvatore Carta. State -of-the-Art in Group Recommendation and New Approaches for Automatic Identification of Groups. In DART, pages 1–20, 2011

    2011

  3. [10]

    Attention-based Group Recommendation

    Tran Dang Quang Vinh, Tuan-Anh Nguyen Pham, Gao Cong, Xiao-Li Li. Attention-based Group Recommendation. CoRR abs/1804.04327, 2018

    2018 arXiv

  4. [11]

    Attentive Group Recommendation

    Da Cao, Xiangnan He, Lianhai Miao, Yahui An, Chao Yang, Richang Hong. Attentive Group Recommendation. In SIGIR, pages 645 -654, 2018

    2018

  5. [12]

    Quintarelli, E

    E. Quintarelli, E. Rabosio, and L. Tanca. Recommending new items to ephemeral groups using contextual user influ ence. In RecSys, pa ges 285–292, 2016

    2016

  6. [13]

    Group recommender systems: Combining individuals models// Ricci F, Rokach L, Shapira B, Kantor P B eds

    Masthoff J. Group recommender systems: Combining individuals models// Ricci F, Rokach L, Shapira B, Kantor P B eds. Recommender Systems Handbook. Berlin: Springer-Verlag, 677-702, 2011

    2011

  7. [14]

    Zhang and B

    Y. Zhang and B. Zuo. Social identity theory and it s development. Advances in Psychological Science, 2006, 14(03): 475-480

    2006

  8. [15]

    B. Cui. From Two-Step to N-Step: A Survey of Two-step flow theory. Journalism Communication, 2014, 11(05): 163-164

    2014

  9. [17]

    McCarthy, M

    K. McCarthy, M. Salamó , L. Coyle, L. McGinty, B. Smyth, and P. Nixon. Cats. A synchronous approach to collaborative group recommendation. In FLAIRS, pages 86–91, 2006

    2006

  10. [18]

    J. F. McCarthy. Pock et restaurant finder: A situated recommender systems for groups. In Proceeding of Workshop on Mobile Ad -Hoc Communication at the 2002 ACM Conference on Human Factors in Computer Systems, 2002

    2002

  11. [19]

    Crossen, J

    A. Crossen, J. Budzik, and K. J. Hammond. Flytrap: intelligent group music recommendation. In IUI, pages 184–185, 2002

    2002

  12. [20]

    Khoshgoftaar

    Xiaoyuan Su, Taghi M. Khoshgoftaar. A Survey of Collaborative Filtering Techniques. Adv. Artificial Intellegence 2009: 421425:1 - 421425:19 (2009)

    2009

  13. [21]

    J. F. McCarthy and T. D. Anagnost. Musicfx: An arbite r of group preferences for computer aupported collaborative workouts. In CSCW, pages 363–372, 1998

    1998

  14. [22]

    Z. Yu, X. Zhou, Y. Hao, and J. Gu. Tv program recommendation for multiple viewers based on user profile merging. User Model. User - Adapt. Interact., 16(1):63–82, 2006

    2006

  15. [23]

    O’Connor, D

    M. O’Connor, D. Cosley, J. A. Konstan, and J. Riedl. Polylens. A recommender system for groups of user. In ECSCW, pages 199 –218, 2001

    2001

  16. [24]

    Pizzutilo, B

    S. Pizzutilo, B. De Carolis, G. Cozzolongo, and F. Ambruoso. Group modeling in a public space: methods, techni ques, experiences. In WSEAS, pages 175–180, 2005

    2005

  17. [25]

    Users’ satisfaction in recommendation systems for groups: an approach based on noncooperative games

    Lucas Augusto Montalvã o Costa Carvalho and Hendrik Teixeira Macedo. Users’ satisfaction in recommendation systems for groups: an approach based on noncooperative games. In WWW, pages 951 –958, 2013

    2013

  18. [26]

    Exploring personal impact for group recommendation

    Xingjie Liu, Yuan Tian, Mao Ye, and Wang -Chien Lee. Exploring personal impact for group recommendation. In CIKM, pages 674–683, 2012

    2012

  19. [27]

    Effective Multi -Query Expansions: Collaborative Deep Networks for Robust Landmark Retrieval

    Y Wang, X Lin, L Wu, et al. Effective Multi -Query Expansions: Collaborative Deep Networks for Robust Landmark Retrieval. IEEE Transactions on Image Processing 26 (3), 1393-1404, 2017

    2017

  20. [28]

    Cycle -Consistent Deep Generative Hashing for Cross-Modal Retrieval

    L Wu, Y Wang, L Shao. Cycle -Consistent Deep Generative Hashing for Cross-Modal Retrieval. IEEE Transactions on Image Processing 28 (4), 1602-1612, 2019

    2019

  21. [29]

    Deep learning

    LeCun Y, Bengio Y, Hinton G. Deep learning. Natur e, 2015, 521(7553): 436-444

    2015

  22. [30]

    24(11):3939-3949, 2015

    Y Wang, X Lin, L Wu et al., Robust Subspace Clustering for Multi - view Data by Exploiting Correlation Consensus, IEEE Transactions on Image Processing. 24(11):3939-3949, 2015

    2015

  23. [31]

    In IJCAI, pages 2153-2159, 2016

    Y Wang et al., Iterative Views Agreement: An Iterative Low-Rank based Structured Optimization Method to Multi -View Spectral Clustering. In IJCAI, pages 2153-2159, 2016

    2016

  24. [32]

    Cross -media analysis and reasoning: Advance and direction

    Peng Y, Zhu W, Zhao Y, et al. Cross -media analysis and reasoning: Advance and direction. Frontiers of Information Technology & Electronic Engineering, 2017,18(1): 44-57

    2017

  25. [33]

    Deep Attention-based Spatially Recursive Networks for Fine-Grained Visual Recognition

    L Wu, Y Wang, X Li, J Gao. Deep Attention-based Spatially Recursive Networks for Fine-Grained Visual Recognition. IEEE Transactions on Cybernetics 49 (5), 1791-1802, 2019

    2019

  26. [34]

    X. Geng, H. Zhang, J. Bian, and T. Chua. Learning image and user features for recommendation in social networks. In ICCV, pages 4274– 4282, 2015

    2015

  27. [35]

    W. Yu, H. Zhang, X. He, X. Chen, L. Xiong, and Z. Qin. Aesthetic - based clothing recommendation. In WWW, pages 649–658, 2018

    2018

  28. [36]

    Embedding-based News Recommendation for Millions of Users

    Shumpei Okura, Yukihiro Tagami, Shingo Ono, and Akira Tajima. Embedding-based News Recommendation for Millions of Users. In SIGKDD, pages 1933–1942, 2017

    1933

  29. [37]

    X. He, L. Liao, H. Zhang, L. Nie, X. Hu, and T. -S. Chua. Neural collaborative filtering. In WWW, pages 173–182, 2017

    2017

  30. [38]

    X. Wang, X. He, L. Nie, and T. -S. Chua. Item silk road: Recommending items from information domains to social users. In SIGIR, pages 185–194, 2017

    2017

  31. [39]

    Cheng, Y

    Z. Cheng, Y. Ding, X. He, L. Zhu, X. Song, and M. Kankanhalli. A3NCF: An adaptive aspect a ttention model for rating prediction. In IJCAI, pages 3748-3754, 2018

    2018

  32. [40]

    Ark V , S. D . Group Dynamics: The Psychology of Small Group Behavior. Journal of Music Therapy, 1972, 9(4):203-204

    1972

  33. [41]

    Personal influence: the part played by people in the flow of mass communications

    Elihu Katz. Personal influence: the part played by people in the flow of mass communications. Glencoe, Illinois, 1955

    1955

  34. [42]

    Diffusion of innovations, Simon and Schuster, 1962

    Everett Rogers. Diffusion of innovations, Simon and Schuster, 1962

    1962

  35. [43]

    Online Social Network Analysis

    Binxing Fang. Online Social Network Analysis. 192-211, 2014

    2014

  36. [44]

    BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

    Jacob Devlin, Ming -Wei Chang, Kenton Lee, Kristina Toutanova . BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. arXiv:1810.04805, 2018

    2018 arXiv

  37. [45]

    Deep Modeling of Group Preferences for Group -Based Recommendation

    Liang Hu, Jian Cao, Guandong Xu, Longbing Cao, Zhiping Gu, Wei Cao. Deep Modeling of Group Preferences for Group -Based Recommendation. In AAAI, pages 1861-1867, 2014

    2014

  38. [47]

    Self -Attentive Sequential Recommendation

    Kang W C , Mcauley J . Self -Attentive Sequential Recommendation. In ICDM, pages 197-206, 2018

    2018

  39. [48]

    D - CARS:A Declarative Co ntext-Aware Recommender System

    Rosni Lumbantoruan, Xiangmin Zhou, Yongli Ren, Zhifeng Bao. D - CARS:A Declarative Co ntext-Aware Recommender System. In ICDM, pages 1152-1157, 2018

    2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.