Generating Bags of Words from the Sums of Their Word Embeddings

White, Lyndon; Togneri, Roberto; Liu, Wei; Bennamoun, Mohammed

doi:10.1007/978-3-319-75477-2_5

Cited by 11 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is easy to see that unigram embeddings can be written as a linear system Ax where A ∈ R n×p contains v wi in the i th column and x ∈ R p is the bag-of-words vector which counts the number of occurances of words in a text. This application is considered in text processing applications to obtain the original text document given the unigram embeddings [64].…”

Section: Basis Pursuit For Recovering Bag-of-words Of Text Documentsmentioning

confidence: 99%

An adaptive primal-dual framework for nonsmooth convex minimization

et al. 2019

View full text Add to dashboard Cite

We propose a new self-adaptive, double-loop smoothing algorithm to solve composite, nonsmooth, and constrained convex optimization problems. Our algorithm is based on Nesterov's smoothing technique via general Bregman distance functions. It selfadaptively selects the number of iterations in the inner loop to achieve a desired complexity bound without requiring the accuracy a priori as in variants of Augmented Lagrangian methods (ALM). We prove O 1 k -convergence rate on the last iterate of the outer sequence for both unconstrained and constrained settings in contrast to ergodic rates which are common in ALM as well as alternating direction method-of-multipliers literature. Compared to existing inexact ALM or quadratic penalty methods, our analysis does not rely on the worst-case bounds of the subproblem solved by the inner loop. Therefore, our algorithm can be viewed as a restarting technique applied to the ASGARD method in [58] but with rigorous theoretical guarantees or as an inexact ALM with explicit inner loop termination rules and adaptive parameters. Our algorithm only requires to initialize the parameters once, and automatically update them during the iteration process without tuning. We illustrate the superiority of our methods via several examples as compared to the state-of-the-art.

show abstract

Section: Basis Pursuit For Recovering Bag-of-words Of Text Documentsmentioning

confidence: 99%

An adaptive primal-dual framework for nonsmooth convex minimization

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The Sum of Word Embeddings (SoWE) of a reviewer's review text is used as a semantic representation of each reviewer. SoWE refers to the simple linear function of aggregating the embeddings of words to represent a sentence, a reviewer, or a group, shown to be effective in different domains, such as Ren et al [19], White et al [20], and Lyndon et al [21]. Therefore, to obtain a SoWE for a sentence we applied an element-wise average to WEs of a sentence:…”

Section: A Spatial Modelingmentioning

confidence: 99%

Spatio-Temporal Graph Representation Learning for Fraudster Group Detection

Shehnepoor,

Togneri,

Liu

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Motivated by potential financial gain, companies may hire fraudster groups to write fake reviews to either demote competitors or promote their own businesses. Such groups are considerably more successful in misleading customers, as people are more likely to be influenced by the opinion of a large group. To detect such groups, a common model is to represent fraudster groups static networks, consequently overlooking the longitudinal behavior of a reviewer thus the dynamics of co-review relations among reviewers in a group. Hence, these approaches are incapable of excluding outlier reviewers, which are fraudsters intentionally camouflaging themselves in a group and genuine reviewers happen to co-review in fraudster groups. To address this issue, in this work, we propose to first capitalise on the effectiveness of the HIN-RNN in both reviewers' representation learning while capturing the collaboration between reviewers, we first utilize the HIN-RNN to model the co-review relations of reviewers in a group in a fixed time window of 28 days. We refer to this as spatial relation learning representation to signify the generalisability of this work to other networked scenarios. Then we use an RNN on the spatial relations to predict the spatiotemporal relations of reviewers in the group. In the third step, a Graph Convolution Network (GCN) refines the reviewers' vector representations using these predicted relations. These refined representations are then used to remove outlier reviewers. The average of the remaining reviewers' representation is then fed to a simple fully connected layer to predict if the group is a fraudster group or not. Exhaustive experiments of the proposed approach showed a 5% (4%), 12% (5%), 12% (5%) improvement over three of the most recent approaches on precision, recall, and F1-value over the Yelp (Amazon) dataset, respectively.

show abstract

“…Therefore, we use the SoWE at two different levels in this work: reviewer-level and group-level. Note, it is possible to use either sentence or document embedding techniques such as sec2vec, we choose SoWE due to its simplicity and proven efficiency in various domains [16], [20], [21]. The sentence-level embeddings are finetuned by a CNN with max-pooling.…”

Section: A Reviewer Representation Extractionmentioning

confidence: 99%

HIN-RNN: A Graph Representation Learning Neural Network for Fraudster Group Detection With No Handcrafted Features

Shehnepoor,

Togneri,

Liu

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Social reviews are indispensable resources for modern consumers' decision making. For financial gain, companies pay fraudsters preferably in groups to demote or promote products and services since consumers are more likely to be misled by a large amount of similar reviews from groups. Recent approaches on fraudster group detection employed handcrafted features of group behaviors without considering the semantic relation between reviews from the reviewers in a group. In this paper, we propose the first neural approach, HIN-RNN, a Heterogeneous Information Network (HIN) Compatible RNN for fraudster group detection that requires no handcrafted features. HIN-RNN provides a unifying architecture for representation learning of each reviewer, with the initial vector as the sum of word embeddings of all review text written by the same reviewer, concatenated by the ratio of negative reviews. Given a co-review network representing reviewers who have reviewed the same items with the same ratings and the reviewers' vector representation, a collaboration matrix is acquired through HIN-RNN training. The proposed approach is confirmed to be effective with marked improvement over state-of-the-art approaches on both the Yelp (22% and 12% in terms of recall and F1-value, respectively) and Amazon (4% and 2% in terms of recall and F1-value, respectively) datasets.

show abstract

Generating Bags of Words from the Sums of Their Word Embeddings

Cited by 11 publications

References 10 publications

An adaptive primal-dual framework for nonsmooth convex minimization

An adaptive primal-dual framework for nonsmooth convex minimization

Spatio-Temporal Graph Representation Learning for Fraudster Group Detection

HIN-RNN: A Graph Representation Learning Neural Network for Fraudster Group Detection With No Handcrafted Features

Contact Info

Product

Resources

About