Tag2Vec: Learning Tag Representations in Tag Networks

Wang, Junshan; Lu, Zhicong; Song, Guojia; Fan, Yue; Du, Lun; Lin, Wei

doi:10.1145/3308558.3313622

Cited by 17 publications

(10 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, adversarial perturbations on model parameters demonstrates an entirely new paradigm of adversarial training and shows great potential in promoting model effectiveness, especially for those solving transductive embedding problems. A reasonable explanation is in urgent need to exhibit why and where APP can work well, and thus we are able to determine whether APP can be generalized to other similar scenarios such as word embedding [25,30], tag embedding [45,47], table embedding [13,18,52] and broader domains.…”

Section: Introductionmentioning

confidence: 99%

“…Comparison of AS and NM on Cora, Wiki and Citeseer, revealing the similarity of APP and Momentum in practice. 45. 40.45 42.76 43.78 46.28 47.31 48.00 48.54 50.55 50.97 55.11 56.86 58.29 58.71 59.55 60.83 60.32 61.68 Citeseer_NM 51.56 36.28 40.43 43.38 44.46 46.88 47.36 48.74 49.02 50.71 50.79 55.07 57.47 58.41 58.93 59.82 60.59 59.98 59.69 3: Accuracy(%) of multi-class classification on Cora .57 63.54 66.75 69.43 70.54 70.82 71.77 72.66 73.60 74.43 77.75 79.30 80.50 80.89 82.31 82.12 82.32 83.58 LINE 66.02 50.93 56.16 60.55 62.14 63.89 63.46 65.09 65.07 66.18 66.53 68.73 70.15 70.31 71.07 70.93 72.20 72.16 72.80 node2vec 74.71 59.18 64.03 66.94 69.20 70.57 71.00 71.45 72.60 73.33 74.16 77.57 79.93 80.84 81.82 82.14 82.85 83.03 84.13 GF 41.35 20.79 29.76 33.65 34.68 36.13 37.65 39.01 38.93 40.28 40.53 44.85 46.80 48.45 49.75 49.63 50.23 51.75 51.40 GraRep 71.45 55.01 60.85 66.12 68.09 67.26 70.77 70.81 71.27 71.47 72.73 74.42 75.68 75.70 76.13 78.13 77.28 77.43 76.97 GraphSage 53.62 31.51 38.79 44.29 45.07 47.29 49.77 50.43 51.77 53.92 54.18 58.24 59.59 62.07 62.89 63.55 63.79 62.88 65.09 AdvTNE 75.14 58.95 65.18 67.21 68.62 71.19 71.45 72.41 73.39 73.81 75.15 77.88 79.94 81.03 82.22 82.97 83.11 83.71 84.24 Cleora 70.99 54.29 60.97 62.62 65.68 68.51 68.65 68.82 71.17 71.04 72.08 73.85 75.62 76.57 76.82 76.98 77.48 77.90 78.75 EATNE 76.93 60.30 66.28 70.56 72.43 74.96 74.60 75.44 76.64 77.33 77.55 80.12 81.09 81.57 82.47 82.90 83.06 83.47 84.02…”

mentioning

confidence: 99%

See 1 more Smart Citation

Understanding and Improvement of Adversarial Training for Network Embedding from an Optimization Perspective

Chen

Gao

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

Self Cite

View full text Add to dashboard Cite

Network Embedding aims to learn a function mapping the nodes to Euclidean space contribute to multiple learning analysis tasks on networks. However, both the noisy information behind the realworld networks and the overfitting problem negatively impact the quality of embedding vectors. To tackle these problems, researchers utilize Adversarial Perturbations on Parameters (APP) and achieve state-of-the-art performance. Unlike the mainstream methods introducing perturbations on the network structure or the data feature, Adversarial Training for Network Embedding (AdvTNE) adopts APP to directly perturb the model parameters, thus providing a new chance to understand the mechanism behind it. In this paper, we explain APP theoretically from an optimization perspective. Considering the Power-law property of networks and the optimization objective, we analyze the reason for its remarkable results on network embedding. Based on the above analysis and the Sigmoid saturation region problem, we propose a new Sine-base activation to enhance the performance of AdvTNE. We conduct extensive experiments on four real networks to validate the effectiveness of our method in node classification and link prediction. The results demonstrate that our method is competitive with state-of-the-art methods. CCS CONCEPTS• Networks → Topology analysis and generation; Physical topologies; • Information systems → Network data models; • Mathematics of computing → Mathematical optimization.

show abstract

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

Understanding and Improvement of Adversarial Training for Network Embedding from an Optimization Perspective

Chen

Gao

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

“…† Corresponding Authors. 1 CIKM19 Best Research Paper Runner-up Award found in the social network of animals, and category hierarchy is widely used in lots of e-commerce sites (e.g., Alibaba, Amazon and Rakuten Ichiba) [11,12,13]. "Women's Fashion → Tops → Sweaters → Longsleeved knit → Crew neck" is an example of such an organization.…”

Section: Introductionsmentioning

confidence: 99%

Fast Hierarchy Preserving Graph Embedding via Subspace Constraints

Chen

Wang

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Hierarchy preserving network embedding is a method that project nodes into feature space by preserving the hierarchy property of networks. Recently, researches on network representation have considerably profited from taking hierarchy into consideration. Among these works, SpaceNE 1 [1] stands out by preserving hierarchy with the help of subspace constraints on the hierarchy subspace system. However, like all other hierarchy preserving network embedding methods, SpaceNE is time-consuming and cannot generalize to new nodes. In this paper, we propose an inductive method, FastHGE, to learn node representations more efficiently and generalize to new nodes more easily. Empirically, the experiment of node classification demonstrates that the convergence speed of FastHGE is increased by 30 times in the case of the same accuracy with SpaceNE.

show abstract

“…They mainly rely on traditional information retrieval (IR) techniques such as keyword matching [13] or a combination of text similarity and Application Program Interface (API) matching [14]. Recently, many works have taken steps to apply deep learning methods [3,8,18,20,22] to code search [2, 4, 5, 7, 10-12, 17, 19, 21, 23, 24], using neural networks to capture deep and semantic correlations between natural language queries and code snippets, and have achieved promising performance improvements. These methods employ various types of model structures, including sequential models [2,4,5,7,10,17,21,23,24], graph models [6,12], and transformers [4].…”

Section: Introductionmentioning

confidence: 99%

Is a Single Model Enough? MuCoS: A Multi-Model Ensemble Learning for Semantic Code Search

Du,

Shi,

Wang

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Recently, deep learning methods have become mainstream in code search since they do better at capturing semantic correlations between code snippets and search queries and have promising performance. However, code snippets have diverse information from different dimensions, such as business logic, specific algorithm, and hardware communication, so it is hard for a single code representation module to cover all the perspectives. On the other hand, as a specific query may focus on one or several perspectives, it is difficult for a single query representation module to represent different user intents. In this paper, we propose MuCoS, a multimodel ensemble learning architecture for semantic code search. It combines several individual learners, each of which emphasizes a specific perspective of code snippets. We train the individual learners on different datasets which contain different perspectives of code information, and we use a data augmentation strategy to get these different datasets. Then we ensemble the learners to capture comprehensive features of code snippets. The experiments show that MuCoS has better results than the existing state-of-the-art methods. CCS CONCEPTS• Software and its engineering → Reusability; Search-based software engineering; • Information systems → Novelty in information retrieval.

show abstract

Tag2Vec: Learning Tag Representations in Tag Networks

Cited by 17 publications

References 29 publications

Understanding and Improvement of Adversarial Training for Network Embedding from an Optimization Perspective

Understanding and Improvement of Adversarial Training for Network Embedding from an Optimization Perspective

Fast Hierarchy Preserving Graph Embedding via Subspace Constraints

Is a Single Model Enough? MuCoS: A Multi-Model Ensemble Learning for Semantic Code Search

Contact Info

Product

Resources

About