Shallow Reading with Deep Learning: Predicting Popularity of Online Content Using only Its Title

Trzciński, Tomasz Piotr; Wołk, Krzysztof; Marasek, Krzysztof; Rokita, Przemyslaw Stefan

doi:10.1007/978-3-319-60438-1_14

Cited by 20 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we briefly review several related works on video popularity prediction. Considerable efforts are dedicated to exploring the popularity prediction of items such as text [ 21 , 22 ] , images [ 23 , 24 ] , and videos [ 15 – 17 , 25 , 26 ] because of their potential value in business [ 27 , 28 ] .…”

Section: Related Workmentioning

confidence: 99%

Online Video Popularity Regression Prediction Model with Multichannel Dynamic Scheduling Based on User Behavior

Qiao

Pang

Wang

et al. 2021

Chinese J of Electronics

View full text Add to dashboard Cite

Popularity prediction of online video is widely used in many different scenarios. It can not only help video service providers to schedule video web sites, but also bring considerable profits on investment for both providers and advertisers if popularity of online video is predicted accurately. However, online video popularity prediction still cannot have a satisfactory result, due to the complexity of many crucial factors especially of video distribution network. In this article, we extract seven factors from huge amounts of data about user behavior, establishing a new multiple linear regression model to initially predict online video popularity. After that, a multichannel video popularity dynamic scheduling model is proposed to schedule videos on which channel and what time to be broadcast, according to its popularity predicted by multiple linear regression model, ensuring that maximum the sum value of online video popularity of each channel. Experimental results on dataset obtained from Sohu Video, a video service provider in China, and real‐world video flow in Sohu Video demonstrate that the proposed model is robust and has promising performance in predicting online video popularity, which is helpful for video service providers to schedule videos on web sites effectively in the future.

show abstract

Section: Related Workmentioning

confidence: 99%

Online Video Popularity Regression Prediction Model with Multichannel Dynamic Scheduling Based on User Behavior

Qiao

Pang

Wang

et al. 2021

Chinese J of Electronics

View full text Add to dashboard Cite

show abstract

“…Research on text popularity prediction [27][28][29][30][31][32] utilizes different features, such as textual and social features, extracted from the input text messages posted on various social media platforms, such as Twitter and Facebook, to build a predictive model and then apply it to test messages to automatically predict whether the message will be reposted, liked, commented on, and shared in the future. The features in the text popularity prediction task are social features, which show the social interaction of the user (i.e., number of followers, friends, favorites, etc.)…”

Section: Text Popularity Predictionmentioning

confidence: 99%

“…They first predicted whether the social multimedia would be reshaped by framing the problem as a binary reshare classification problem and then predicted the popularity scores of respective multimedia using multiclass classifiers. Work by [31] utilized a deep learning model (long short term memory) to predict the popularity of news articles using their title information while [32] used a Bayesian approach for retweets prediction.…”

Section: Text Popularity Predictionmentioning

confidence: 99%

Exploring the Power of Multimodal Features for Predicting the Popularity of Social Media Image in a Tourist Destination

Gupta

Jung

Yoo

2020

MTI

View full text Add to dashboard Cite

Social media platforms are widely used nowadays by various businesses to promote their products and services through multimedia content. Instagram is one of those platforms, which is used not only by companies to promote their products but also by local governments to promote tourist destinations. Predicting the popularity of the promotional tourist destination images helps marketers to plan strategically. However, given the abundance of images posted on Instagram daily, identifying the factors that determine the popularity of an image is a big challenge, due to informal and noisy visual content, frequent content evolution, a lack of explicit visual elements, and people’s informal behavior in liking, commenting on, and viewing the images. We present an approach to identify the factors most responsible for the popularity of tourist destinations-related images on Instagram. Our approach provides a proof of concept for an artificial intelligence (AI)-based real-time content management system, which will help to promote a tourist destination. The experiments on a collection of posts crawled from the official Instagram account of Jeju Island, which is one of the most popular tourist destinations in Korea, show that the recency of the post is the most important predictor of the number of likes and comments it will receive. Moreover, the combination of visual content and context features is an excellent predictor of popularity. The number of likes and comments are found to be complementary to each other for predicting image popularity.

show abstract

“…The module's general prediction model used to predict TV drama views. In [39], a bi-directional long-term short-term memory neural network (BiLSTM) was proposed to predict the prevalence of online content. Studied in video and news texts, data sets have shown that deep network performance is greatly improved.…”

Section: Related Workmentioning

confidence: 99%

Deep Learning Based Proactive Caching for Effective WSN-Enabled Vision Applications

et al. 2019

View full text Add to dashboard Cite

Wireless Sensor Networks (WSNs) have a wide range of applications scenarios in computer vision, from pedestrian detection to robotic visual navigation. In response to the growing visual data services in WSNs, we propose a proactive caching strategy based on Stacked Sparse Autoencoder (SSAE) to predict content popularity (PCDS2AW). Firstly, based on Software Defined Network (SDN) and Network Function Virtualization (NFV) technologies, a distributed deep learning network SSAE is constructed in the sink nodes and control nodes of the WSN network. Then, the SSAE network structure parameters and network model parameters are optimized through training. The proactive cache strategy implementation procedure is divided into four steps. (1) The SDN controller is responsible for dynamically collecting user request data package information in the WSNs network. (2) The SSAEs predicts the packet popularity based on the SDN controller obtaining user request data. (3) The SDN controller generates a corresponding proactive cache strategy according to the popularity prediction result. (4) Implement the proactive caching strategy at the WSNs cache node. In the simulation, we compare the influence of spatiotemporal data on the SSAE network structure. Compared with the classic caching strategy Hash + LRU, Betw + LRU, and classic prediction algorithms SVM and BPNN, the proposed PCDS2AW proactive caching strategy can significantly improve WSN performance.

show abstract

Shallow Reading with Deep Learning: Predicting Popularity of Online Content Using only Its Title

Cited by 20 publications

References 15 publications

Online Video Popularity Regression Prediction Model with Multichannel Dynamic Scheduling Based on User Behavior

Online Video Popularity Regression Prediction Model with Multichannel Dynamic Scheduling Based on User Behavior

Exploring the Power of Multimodal Features for Predicting the Popularity of Social Media Image in a Tourist Destination

Deep Learning Based Proactive Caching for Effective WSN-Enabled Vision Applications

Contact Info

Product

Resources

About