Dynamic embeddings for user profiling in twitter

Shangsong Liang; Zhaochun Ren; Xiangliang Zhang; Evangelos Kanoulas

doi:10.1145/3219819.3220043

Dynamic embeddings for user profiling in twitter

Shangsong Liang, Zhaochun Ren, Xiangliang Zhang, Evangelos Kanoulas

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

73 Scopus citations

Abstract

In this paper, we study the problem of dynamic user profiling in Twitter. We address the problem by proposing a dynamic user and word embedding model (DUWE), a scalable black-box variational inference algorithm, and a streaming keyword diversification model (SKDM). DUWE dynamically tracks the semantic representations of users and words over time and models their embeddings in the same space so that their similarities can be effectively measured. Our inference algorithm works with a convex objective function that ensures the robustness of the learnt embeddings. SKDM aims at retrieving top-K relevant and diversified keywords to profile users' dynamic interests. Experiments on a Twitter dataset demonstrate that our proposed embedding algorithms outperform state-of-the-art non-dynamic and dynamic embedding and topic models.

Original language	English (US)
Title of host publication	KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Publisher	Association for Computing Machinery
Pages	1764-1773
Number of pages	10
ISBN (Print)	9781450355520
DOIs	https://doi.org/10.1145/3219819.3220043
State	Published - Jul 19 2018
Event	24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018 - London, United Kingdom Duration: Aug 19 2018 → Aug 23 2018

Publication series

Name	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Conference

Conference	24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018
Country/Territory	United Kingdom
City	London
Period	08/19/18 → 08/23/18

Bibliographical note

Publisher Copyright:
© 2018 Copyright held by the owner/author(s). Publication rights licensed to ACM.

Keywords

Dynamic model
Profiling
Word embeddings

ASJC Scopus subject areas

Software
Information Systems

Access to Document

10.1145/3219819.3220043

Cite this

Liang, S., Ren, Z., Zhang, X., & Kanoulas, E. (2018). Dynamic embeddings for user profiling in twitter. In KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1764-1773). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery. https://doi.org/10.1145/3219819.3220043

Liang, Shangsong ; Ren, Zhaochun ; Zhang, Xiangliang et al. / Dynamic embeddings for user profiling in twitter. KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2018. pp. 1764-1773 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

@inproceedings{693116c866ca4c558230a2add9ad2b67,

title = "Dynamic embeddings for user profiling in twitter",

abstract = "In this paper, we study the problem of dynamic user profiling in Twitter. We address the problem by proposing a dynamic user and word embedding model (DUWE), a scalable black-box variational inference algorithm, and a streaming keyword diversification model (SKDM). DUWE dynamically tracks the semantic representations of users and words over time and models their embeddings in the same space so that their similarities can be effectively measured. Our inference algorithm works with a convex objective function that ensures the robustness of the learnt embeddings. SKDM aims at retrieving top-K relevant and diversified keywords to profile users' dynamic interests. Experiments on a Twitter dataset demonstrate that our proposed embedding algorithms outperform state-of-the-art non-dynamic and dynamic embedding and topic models.",

keywords = "Dynamic model, Profiling, Word embeddings",

author = "Shangsong Liang and Zhaochun Ren and Xiangliang Zhang and Evangelos Kanoulas",

note = "Publisher Copyright: {\textcopyright} 2018 Copyright held by the owner/author(s). Publication rights licensed to ACM.; 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018 ; Conference date: 19-08-2018 Through 23-08-2018",

year = "2018",

month = jul,

day = "19",

doi = "10.1145/3219819.3220043",

language = "English (US)",

isbn = "9781450355520",

series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

publisher = "Association for Computing Machinery",

pages = "1764--1773",

booktitle = "KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

Liang, S, Ren, Z, Zhang, X & Kanoulas, E 2018, Dynamic embeddings for user profiling in twitter. in KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, pp. 1764-1773, 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018, London, United Kingdom, 08/19/18. https://doi.org/10.1145/3219819.3220043

Dynamic embeddings for user profiling in twitter. / Liang, Shangsong; Ren, Zhaochun; Zhang, Xiangliang et al.
KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2018. p. 1764-1773 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Dynamic embeddings for user profiling in twitter

AU - Liang, Shangsong

AU - Ren, Zhaochun

AU - Zhang, Xiangliang

AU - Kanoulas, Evangelos

PY - 2018/7/19

Y1 - 2018/7/19

N2 - In this paper, we study the problem of dynamic user profiling in Twitter. We address the problem by proposing a dynamic user and word embedding model (DUWE), a scalable black-box variational inference algorithm, and a streaming keyword diversification model (SKDM). DUWE dynamically tracks the semantic representations of users and words over time and models their embeddings in the same space so that their similarities can be effectively measured. Our inference algorithm works with a convex objective function that ensures the robustness of the learnt embeddings. SKDM aims at retrieving top-K relevant and diversified keywords to profile users' dynamic interests. Experiments on a Twitter dataset demonstrate that our proposed embedding algorithms outperform state-of-the-art non-dynamic and dynamic embedding and topic models.

AB - In this paper, we study the problem of dynamic user profiling in Twitter. We address the problem by proposing a dynamic user and word embedding model (DUWE), a scalable black-box variational inference algorithm, and a streaming keyword diversification model (SKDM). DUWE dynamically tracks the semantic representations of users and words over time and models their embeddings in the same space so that their similarities can be effectively measured. Our inference algorithm works with a convex objective function that ensures the robustness of the learnt embeddings. SKDM aims at retrieving top-K relevant and diversified keywords to profile users' dynamic interests. Experiments on a Twitter dataset demonstrate that our proposed embedding algorithms outperform state-of-the-art non-dynamic and dynamic embedding and topic models.

KW - Dynamic model

KW - Profiling

KW - Word embeddings

UR - http://www.scopus.com/inward/record.url?scp=85051488450&partnerID=8YFLogxK

U2 - 10.1145/3219819.3220043

DO - 10.1145/3219819.3220043

M3 - Conference contribution

AN - SCOPUS:85051488450

SN - 9781450355520

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 1764

EP - 1773

BT - KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

T2 - 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018

Y2 - 19 August 2018 through 23 August 2018

ER -

Liang S, Ren Z, Zhang X, Kanoulas E. Dynamic embeddings for user profiling in twitter. In KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2018. p. 1764-1773. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). doi: 10.1145/3219819.3220043

Dynamic embeddings for user profiling in twitter

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this