Doc2Cube: Allocating Documents to Text Cube Without Labeled Data

Tao, Fangbo; Zhang, Chao; Chen, Xiusi; Jiang, Meng; Hanratty, Timothy; Kaplan, Lance

doi:10.1109/icdm.2018.00169

Cited by 20 publications

(21 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…WeSTClass (Meng et al, 2018(Meng et al, , 2019c) models class semantics as vMF distributions in the word embedding space and applies a pretrain-refine neural approach to perform text classification under weak supervision. Doc2Cube (Tao et al, 2018) leverages word-document cooccurrences to embed class labels, words and documents in the same space and perform classification by comparing embedding similarity. We adopt the two frameworks and replace the original embedding with the embedding trained by our Joint Skip-Gram and Joint CBOW models.…”

Section: Weakly-supervised Text Classificationmentioning

confidence: 99%

“…Some studies along the embedding line learn word embeddings based on global contexts implicitly . HSMN (Huang et al, 2012 ), PTE (Tang et al, 2015 ), and Doc2Cube (Tao et al, 2018 ) take the average of word embedding in the document as the document representation and encourage similarity between word embedding and document embedding for co-occurred words and documents. However, these methods do not model global contexts explicitly because the document representations are essentially aggregated word representations and thus are not tailored for contextual representations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Word Embedding Learning by Incorporating Local and Global Contexts

Huang

Wang

et al. 2020

Front. Big Data

Self Cite

View full text Add to dashboard Cite

Word embedding has benefited a broad spectrum of text analysis tasks by learning distributed word representations to encode word semantics. Word representations are typically learned by modeling local contexts of words, assuming that words sharing similar surrounding words are semantically close. We argue that local contexts can only partially define word semantics in the unsupervised word embedding learning. Global contexts, referring to the broader semantic units, such as the document or paragraph where the word appears, can capture different aspects of word semantics and complement local contexts. We propose two simple yet effective unsupervised word embedding models that jointly model both local and global contexts to learn word representations. We provide theoretical interpretations of the proposed models to demonstrate how local and global contexts are jointly modeled, assuming a generative relationship between words and contexts. We conduct a thorough evaluation on a wide range of benchmark datasets. Our quantitative analysis and case study show that despite their simplicity, our two proposed models achieve superior performance on word similarity and text classification tasks.

show abstract

Section: Weakly-supervised Text Classificationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unsupervised Word Embedding Learning by Incorporating Local and Global Contexts

Huang

Wang

et al. 2020

Front. Big Data

Self Cite

View full text Add to dashboard Cite

show abstract

“…A. Preliminaries 1) Data Cube Basics: Data cube is widely used to organize multi-dimensional data, such as records in relational databases and documents in text collections [13], [9], [11], [12]. With well-designed cube structures, it can largely boost various downstream data analytics, mining and summarization tasks [8].…”

Section: Problem Formulationmentioning

confidence: 99%

“…Following the design of [12], we can create three cube dimensions based on paper attributes in DBLP: L decade derived from the numerical attribute publication year, L venue derived from the categorical attribute publication venue, and L topic derived from textual attributes like paper titles and abstracts. We assign labels of the venue dimension by publication years; venue labels are assigned by conference or journal names like KDD, TKDE; topic labels are assigned according to a latent phrase-based topic model by labels like Neural networks, Feature selection.…”

Section: Problem Formulationmentioning

confidence: 99%

cube2net: Efficient Query-Specific Network Construction with Data Cube Organization

Yang

2019

2019 International Conference on Data Mining Workshops (ICDMW)

Self Cite

View full text Add to dashboard Cite

Networks are widely used to model objects with interactions and have enabled various downstream applications. However, in the real world, network mining is often done on particular query sets of objects, which does not require the construction and computation of networks including all objects in the datasets. In this work, for the first time, we propose to address the problem of query-specific network construction, to break the efficiency bottlenecks of existing network mining algorithms and facilitate various downstream tasks. To deal with real-world massive networks with complex attributes, we propose to leverage the well-developed data cube technology to organize network objects w.r.t. their essential attributes. An efficient reinforcement learning algorithm is then developed to automatically explore the data cube structures and construct the optimal query-specific networks. With extensive experiments of two classic network mining tasks on different real-world large datasets, we show that our proposed cube2net pipeline is general, and much more effective and efficient in query-specific network construction, compared with other methods without the leverage of data cube or reinforcement learning.

show abstract

“…• Introduction: Motation & Overview • Phrase Mining [2,5,10,11,19] • Named Entity Recognition [3,6,9,12,15,18,20] • Taxonomy Construction [1, 7, 13, 16, 17, 23-25, 29, 30] • Mining Constructed Networks [4,8,14,21,[26][27][28] • System Demos [22,27] • Summary and Future Directions • Question Answering and Discussions…”

Section: Detailed Tutorial Outlinementioning

confidence: 99%

Constructing and Mining Heterogeneous Information Networks from Massive Text

Shang

Shen

Liu

2019

Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

Real-world data exists largely in the form of unstructured texts. A grand challenge on data mining research is to develop effective and scalable methods that may transform unstructured text into structured knowledge. Based on our vision, it is highly beneficial to transform such text into structured heterogeneous information networks, on which actionable knowledge can be generated based on the user's need. In this tutorial, we provide a comprehensive overview on recent research and development in this direction. First, we introduce a series of effective methods that construct heterogeneous information networks from massive, domain-specific text corpora. Then we discuss methods that mine such text-rich networks based on the user's need. Specifically, we focus on scalable, effective, weakly supervised, language-agnostic methods that work on various kinds of text. We further demonstrate, on real datasets (including news articles, scientific publications, and product reviews), how information networks can be constructed and how they can assist further exploratory analysis.

show abstract

Doc2Cube: Allocating Documents to Text Cube Without Labeled Data

Cited by 20 publications

References 12 publications

Unsupervised Word Embedding Learning by Incorporating Local and Global Contexts

Unsupervised Word Embedding Learning by Incorporating Local and Global Contexts

cube2net: Efficient Query-Specific Network Construction with Data Cube Organization

Constructing and Mining Heterogeneous Information Networks from Massive Text

Contact Info

Product

Resources

About