Uncovering Suspicious Activity From Partially Paired and Incomplete Multimodal Data

Chiu, Carter; Zhan, Justin; Zhan, F. Benjamin

doi:10.1109/access.2017.2726078

Cited by 19 publications

(6 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each user, the ID, IP, time of the written reviews and retweet comment was used to find fraud reviews on Twitter. Chiu et al [2017] also utilized user behavior analysis to find suspicious activity, this time through multimodal data. Multimodal data refers to data gathered from different sources such as security devices, networks, servers, and applications.…”

Section: Single Usermentioning

confidence: 99%

“…Multimodal data refers to data gathered from different sources such as security devices, networks, servers, and applications. Chiu et al [2017] employed three modalities from Data Modality [Baltrušaitis et al 2017] referring to data derived from different layers of the review writing process (from IP address at the lowest layer to review text at the highest layer). Twitter messages obtained through Streaming API, certain user data (e.g., User ID, Tweet ID, and Time), Network Information (e.g., IP Address), and Device Information (e.g., GPS position).…”

Section: Single Usermentioning

confidence: 99%

“…The matrix in each time step represents the the user ID, review ID, and the date of the written review as data modalities. Chiu et al [2017] provided an intuition of how modeling the behaviors of the user through time helps to improve the modeling of user behavior.…”

Section: Single Usermentioning

confidence: 99%

“…It is worth mentioning that some platforms provide privileges (e.g. APIs) for the crawlers to collect data from different data modalities, also referred to as "Multi-modal" data [Chiu et al 2017]. For example, ; Wen et al [2020] collected data from the Tencent platform which includes IP and different physical layers alongside information from the application layer.…”

Section: Datamentioning

confidence: 99%

See 3 more Smart Citations

Social Fraud Detection Review: Methods, Challenges and Analysis

Shehnepoor¹,

Togneri²,

Liu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Social reviews have dominated the web and become a plausible source of product information. People and businesses use such information for decision-making. Businesses also make use of social information to spread fake information using a single user, groups of users, or a bot trained to generate fraudulent content. Many studies proposed approaches based on user behaviors and review text to address the challenges of fraud detection. To provide an exhaustive literature review, social fraud detection is reviewed using a framework that considers three key components: the review itself, the user who carries out the review, and the item being reviewed. As features are extracted for the component representation, a feature-wise review is provided based on behavioral, text-based features and their combination. With this framework, a comprehensive overview of approaches is presented including supervised, semi-supervised, and unsupervised learning. The supervised approaches for fraud detection are introduced and categorized into two sub-categories; classical, and deep learning. The lack of labeled datasets is explained and potential solutions are suggested. To help new researchers in the area develop a better understanding, a topic analysis and an overview of future directions is provided in each step of the proposed systematic framework.

show abstract

Section: Single Usermentioning

confidence: 99%

Section: Single Usermentioning

confidence: 99%

Section: Single Usermentioning

confidence: 99%

Section: Datamentioning

confidence: 99%

See 2 more Smart Citations

Social Fraud Detection Review: Methods, Challenges and Analysis

Shehnepoor¹,

Togneri²,

Liu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In another study, [8], the question of identifying suspect blocks in multi-modal data is considered, although the data is incomplete. The authors proposed a method to measure block suspicion in a multi-modal tensor and demonstrated the satisfaction with axioms which delineate an effective metric as advanced by past research and also with the additional axiom.…”

Section: Literature Reviewmentioning

confidence: 99%

Suspicious Activity Detection of Twitter and Facebook using Sentimental Analysis

Mansoori¹,

Almansoori²,

Alshamsi³

et al. 2020

TEM Journal

View full text Add to dashboard Cite

The purpose of this study is to evaluate the criminal behavior on the social media platforms and to classify the gathered data effectively as negative, positive, or neutral in order to identify a suspect. In this study, data was collected from two platforms, Twitter and Facebook, resulting in the creation of two datasets. The following findings have been pointed out from this study: Initially, VADER twitter sentimental analysis showed that out of 5000 tweets 50.8% people shared a neutral opinion, 39.2% shared negative opinion and only 9.9% showed positive opinion. Secondly, on Facebook, the majority of people showed a neutral response which is 55.6%, 38.9% shared positive response and only 5.6% shared negative opinion. Thirdly, the score of sentiments and engagement in every post affects the intensities of sentiments.

show abstract

Deep Learning for Link Prediction in Dynamic Networks Using Weak Estimators

Chiu

Zhan

2018

IEEE Access

View full text Add to dashboard Cite

Uncovering Suspicious Activity From Partially Paired and Incomplete Multimodal Data

Cited by 19 publications

References 29 publications

Social Fraud Detection Review: Methods, Challenges and Analysis

Social Fraud Detection Review: Methods, Challenges and Analysis

Suspicious Activity Detection of Twitter and Facebook using Sentimental Analysis

Deep Learning for Link Prediction in Dynamic Networks Using Weak Estimators

Contact Info

Product

Resources

About