Social network advertising is currently one of the most effective advertising types available to promote a product or a brand. The problem discussed in this paper concerns the possibility to ensure that advertising reaches really interested users, and also to prove this. At this aim, we propose the use of Blockchain to store users' interest and to obtain an assertion that a user is interested in a product before the advertising is shown. The proposal has been implemented by a Solidity smart contract in Ethereum and has been shown to be effective and cheap.
{"title":"A System to Enforce User's Preference in OSN Advertising","authors":"G. Lax, A. Russo","doi":"10.1145/3341161.3345618","DOIUrl":"https://doi.org/10.1145/3341161.3345618","url":null,"abstract":"Social network advertising is currently one of the most effective advertising types available to promote a product or a brand. The problem discussed in this paper concerns the possibility to ensure that advertising reaches really interested users, and also to prove this. At this aim, we propose the use of Blockchain to store users' interest and to obtain an assertion that a user is interested in a product before the advertising is shown. The proposal has been implemented by a Solidity smart contract in Ethereum and has been shown to be effective and cheap.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"145 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123261482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
When studying a network, it is often of interest to understand the robustness of that network to noise. Network robustness has been studied in a variety of contexts, examining network properties such as the number of connected components and the lengths of shortest paths. In this work, we present a new network robustness measure, which we refer to as ‘sampling robustness'. The goal of the sampling robustness measure is to quantify the extent to which a network sample collected from a graph with errors is a good representation of a network sample collected from that same graph, but without errors. These errors may be introduced by humans or by the system (e.g., mistakes from the respondents or a bug in an API program), and may affect the performance of a data collection algorithm and the quality of the obtained sample. Thus, when data analysts analyze the sampled network, they may wish to know whether such errors will affect future analysis results. We demonstrate that sampling robustness is dependent on a few easily-computed properties of the network: the leading eigenvalue, average node degree and clustering coefficient. In addition, we introduce regression models for estimating sampling robustness given an obtained sample. As a result, our models can estimate the sampling robustness with MSE < 0.0015 and the model has an R-squared of up to 75%.
{"title":"Measuring the Sampling Robustness of Complex Networks","authors":"K. Areekijseree, S. Soundarajan","doi":"10.1145/3341161.3342873","DOIUrl":"https://doi.org/10.1145/3341161.3342873","url":null,"abstract":"When studying a network, it is often of interest to understand the robustness of that network to noise. Network robustness has been studied in a variety of contexts, examining network properties such as the number of connected components and the lengths of shortest paths. In this work, we present a new network robustness measure, which we refer to as ‘sampling robustness'. The goal of the sampling robustness measure is to quantify the extent to which a network sample collected from a graph with errors is a good representation of a network sample collected from that same graph, but without errors. These errors may be introduced by humans or by the system (e.g., mistakes from the respondents or a bug in an API program), and may affect the performance of a data collection algorithm and the quality of the obtained sample. Thus, when data analysts analyze the sampled network, they may wish to know whether such errors will affect future analysis results. We demonstrate that sampling robustness is dependent on a few easily-computed properties of the network: the leading eigenvalue, average node degree and clustering coefficient. In addition, we introduce regression models for estimating sampling robustness given an obtained sample. As a result, our models can estimate the sampling robustness with MSE < 0.0015 and the model has an R-squared of up to 75%.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129570398","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Twitter user geolocation detection can inform and benefit a range of downstream geospatial tasks such as event and venue recommendation, local search, and crisis planning and response. In this paper, we take into account user shared tweets as well as their social network, and run extensive comparative studies to systematically analyze the impact of a variety of language-based, network-based, and hybrid methods in predicting user geolocation. In particular, we evaluate different text representation methods to construct text views that capture the linguistic signals available in tweets that are specific to and indicative of geographical locations. In addition, we investigate a range of network-based methods, such as embedding approaches and graph neural networks, in predicting user geolocation based on user interaction network. Our findings provide valuable insights into the design of effective and efficient geolocation identification engines. Finally, our best model, called TF-MF, substantially outperforms state-of-the-art approaches under minimal supervision.
{"title":"TF-MF: Improving Multiview Representation for Twitter User Geolocation Prediction","authors":"P. Hamouni, Taraneh Khazaei, Ehsan Amjadian","doi":"10.1145/3341161.3342961","DOIUrl":"https://doi.org/10.1145/3341161.3342961","url":null,"abstract":"Twitter user geolocation detection can inform and benefit a range of downstream geospatial tasks such as event and venue recommendation, local search, and crisis planning and response. In this paper, we take into account user shared tweets as well as their social network, and run extensive comparative studies to systematically analyze the impact of a variety of language-based, network-based, and hybrid methods in predicting user geolocation. In particular, we evaluate different text representation methods to construct text views that capture the linguistic signals available in tweets that are specific to and indicative of geographical locations. In addition, we investigate a range of network-based methods, such as embedding approaches and graph neural networks, in predicting user geolocation based on user interaction network. Our findings provide valuable insights into the design of effective and efficient geolocation identification engines. Finally, our best model, called TF-MF, substantially outperforms state-of-the-art approaches under minimal supervision.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128800624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chia-Wei Chen, Sheng-Chuan Chou, Chang-You Tai, Lun-Wei Ku
As deep learning models are getting popular, upgrading the retrieval-based content recommendation system to the learning-based system is highly demanded. However, efficiency is a critical issue. For article recommendation, an effective neural network which generates a good representation of the article content could prove useful. Hence, we propose PGA-Recommender, a phrase-guided article recommendation model which mimics the process of human behavior - first browsing, then guided by key phrases, and finally aggregating the gleaned information. As this can be performed independently offline, it is thus compatible with current commercial retrieval-based (keyword-based) article recommender systems. A total of six months of real logs - from Apr 2017 to Sep 2017 - were used for experiments. Results show that PGA-Recommender outperforms different state-of-the-art schemes including session-, collaborative filter-, and content-based recommendation models. Moreover, it suggests a diverse mix of articles while maintaining superior performance in terms of both click and view predictions. The results of A/B tests show that simply using the backward version of PGA-Recommender yields 40% greater click-through rates as compared to the retrieval-based system when deployed to a language of which we have zero knowledge.
{"title":"Phrase-Guided Attention Web Article Recommendation for Next Clicks and Views","authors":"Chia-Wei Chen, Sheng-Chuan Chou, Chang-You Tai, Lun-Wei Ku","doi":"10.1145/3341161.3342869","DOIUrl":"https://doi.org/10.1145/3341161.3342869","url":null,"abstract":"As deep learning models are getting popular, upgrading the retrieval-based content recommendation system to the learning-based system is highly demanded. However, efficiency is a critical issue. For article recommendation, an effective neural network which generates a good representation of the article content could prove useful. Hence, we propose PGA-Recommender, a phrase-guided article recommendation model which mimics the process of human behavior - first browsing, then guided by key phrases, and finally aggregating the gleaned information. As this can be performed independently offline, it is thus compatible with current commercial retrieval-based (keyword-based) article recommender systems. A total of six months of real logs - from Apr 2017 to Sep 2017 - were used for experiments. Results show that PGA-Recommender outperforms different state-of-the-art schemes including session-, collaborative filter-, and content-based recommendation models. Moreover, it suggests a diverse mix of articles while maintaining superior performance in terms of both click and view predictions. The results of A/B tests show that simply using the backward version of PGA-Recommender yields 40% greater click-through rates as compared to the retrieval-based system when deployed to a language of which we have zero knowledge.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125310991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Caleb Belth, Fahad Kamran, Donna Tjandra, Danai Koutra
For trajectory data that tend to have beyond first-order (i.e., non-Markovian) dependencies, higher-order networks have been shown to accurately capture details lost with the standard aggregate network representation. At the same time, representation learning has shown success on a wide range of network tasks, removing the need to hand-craft features for these tasks. In this work, we propose a node representation learning framework called EVO or Embedding Variable Orders, which captures non-Markovian dependencies by combining work on higher-order networks with work on node embeddings. We show that EVO outperforms baselines in tasks where high-order dependencies are likely to matter, demonstrating the benefits of considering high-order dependencies in node embeddings. We also provide insights into when it does or does not help to capture these dependencies. To the best of our knowledge, this is the first work on representation learning for higher-order networks.
{"title":"When to Remember Where You Came from: Node Representation Learning in Higher-order Networks","authors":"Caleb Belth, Fahad Kamran, Donna Tjandra, Danai Koutra","doi":"10.1145/3341161.3342911","DOIUrl":"https://doi.org/10.1145/3341161.3342911","url":null,"abstract":"For trajectory data that tend to have beyond first-order (i.e., non-Markovian) dependencies, higher-order networks have been shown to accurately capture details lost with the standard aggregate network representation. At the same time, representation learning has shown success on a wide range of network tasks, removing the need to hand-craft features for these tasks. In this work, we propose a node representation learning framework called EVO or Embedding Variable Orders, which captures non-Markovian dependencies by combining work on higher-order networks with work on node embeddings. We show that EVO outperforms baselines in tasks where high-order dependencies are likely to matter, demonstrating the benefits of considering high-order dependencies in node embeddings. We also provide insights into when it does or does not help to capture these dependencies. To the best of our knowledge, this is the first work on representation learning for higher-order networks.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130157046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ki-young Shin, Woosang Song, Jinhee Kim, Jong-Hyeok Lee
We provide a more optimized model for calculating credibility score of information in SNS. We premeditated two heuristics which using characteristics of the credibility score for each document: (1) Expertise and (2) un-biasedness. Also, we divide the users in SNs into three types: (1) Creator (2) Distributor, and (3) Follower. Our model is designed to calculate Expertise and Un-biasedness for three types of SNs users (Creator, Distributor, and Follower) by using logistic regression model. Our model not only reveals whether the information is ‘accurate and unbiased’, but also investigates the ‘source, distribution channel, and audience’ of the information. We expect our credibility scoring will give answers to the ‘qualitative problem’ our online world is currently facing.
{"title":"News Credibility Scroing: Suggestion of research methodology to determine the reliability of news distributed in SNS","authors":"Ki-young Shin, Woosang Song, Jinhee Kim, Jong-Hyeok Lee","doi":"10.1145/3341161.3343683","DOIUrl":"https://doi.org/10.1145/3341161.3343683","url":null,"abstract":"We provide a more optimized model for calculating credibility score of information in SNS. We premeditated two heuristics which using characteristics of the credibility score for each document: (1) Expertise and (2) un-biasedness. Also, we divide the users in SNs into three types: (1) Creator (2) Distributor, and (3) Follower. Our model is designed to calculate Expertise and Un-biasedness for three types of SNs users (Creator, Distributor, and Follower) by using logistic regression model. Our model not only reveals whether the information is ‘accurate and unbiased’, but also investigates the ‘source, distribution channel, and audience’ of the information. We expect our credibility scoring will give answers to the ‘qualitative problem’ our online world is currently facing.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116269138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Young D. Kwon, Reza Hadi Mogavi, E. Haq, Young D. Kwon, Xiaojuan Ma, Pan Hui
Understanding how much users disclose personal information in Online Social Networks (OSN) has served various scenarios such as maintaining social relationships and customer segmentation. Prior studies on self-disclosure have relied on surveys or users' direct social networks. These approaches, however, cannot represent the whole population nor consider user dynamics at the community level. In this paper, we conduct a quantitative study at different granularities of networks (ego networks and user communities) to understand users' self-disclosing behaviors better. As our first contribution, we characterize users into three types (open, closed, and moderate) based on the Communication Privacy Management theory and extend the analysis of the self-disclosure of users to a large-scale OSN dataset which could represent the entire network structure. As our second contribution, we show that our proposed features of ego networks and positional and structural properties of communities significantly affect self-disclosing behavior. Based on these insights, we present the possible relation between the propensity of the self-disclosure of users and the sociological theory of structural holes, i.e., users at a bridge position can leverage advantages among distinct groups. To the best of our knowledge, our study provides the first attempt to shed light on the self-disclosure of users using the whole network structure, which paves the way to a better understanding of users' self-disclosing behaviors and their relations with overall network structures.
了解用户在在线社交网络(Online Social Networks, OSN)中泄露了多少个人信息，已经服务于维护社会关系和客户细分等各种场景。先前关于自我表露的研究依赖于调查或用户的直接社交网络。然而，这些方法不能代表全体人口，也不能考虑社区一级的用户动态。本文通过对网络(自我网络和用户社区)不同粒度的定量研究，更好地理解用户的自我披露行为。作为我们的第一个贡献，我们基于通信隐私管理理论将用户划分为开放、封闭和适度三种类型，并将用户自我披露的分析扩展到一个可以代表整个网络结构的大规模OSN数据集。作为我们的第二个贡献，我们展示了我们提出的自我网络特征以及社区的位置和结构特性显著影响自我披露行为。基于这些见解，我们提出了用户自我披露倾向与结构漏洞的社会学理论之间的可能关系，即处于桥梁位置的用户可以利用不同群体之间的优势。据我们所知，我们的研究首次尝试揭示了使用整个网络结构的用户自我披露，这为更好地理解用户的自我披露行为及其与整体网络结构的关系铺平了道路。
{"title":"Effects of Ego Networks and Communities on Self-Disclosure in an Online Social Network","authors":"Young D. Kwon, Reza Hadi Mogavi, E. Haq, Young D. Kwon, Xiaojuan Ma, Pan Hui","doi":"10.1145/3341161.3342881","DOIUrl":"https://doi.org/10.1145/3341161.3342881","url":null,"abstract":"Understanding how much users disclose personal information in Online Social Networks (OSN) has served various scenarios such as maintaining social relationships and customer segmentation. Prior studies on self-disclosure have relied on surveys or users' direct social networks. These approaches, however, cannot represent the whole population nor consider user dynamics at the community level. In this paper, we conduct a quantitative study at different granularities of networks (ego networks and user communities) to understand users' self-disclosing behaviors better. As our first contribution, we characterize users into three types (open, closed, and moderate) based on the Communication Privacy Management theory and extend the analysis of the self-disclosure of users to a large-scale OSN dataset which could represent the entire network structure. As our second contribution, we show that our proposed features of ego networks and positional and structural properties of communities significantly affect self-disclosing behavior. Based on these insights, we present the possible relation between the propensity of the self-disclosure of users and the sociological theory of structural holes, i.e., users at a bridge position can leverage advantages among distinct groups. To the best of our knowledge, our study provides the first attempt to shed light on the self-disclosure of users using the whole network structure, which paves the way to a better understanding of users' self-disclosing behaviors and their relations with overall network structures.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115748843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recently, many researchers in recommender systems have realized that encoding user-item interactions based on deep neural networks (DNNs) promotes collaborative-filtering (CF)'s performance. Nonetheless, those DNN-based models' performance is still limited when observed user-item interactions are very less because the training samples distilled from these interactions are critical for deep learning models. To address this problem, we resort to plenty features distilled from knowledge graphs (KGs), to profile users and items precisely and sufficiently rather than observed user-item interactions. In this paper, we propose a knowledge embedding based recommendation framework to alleviate the problem of sparse user-item interactions in recommendation. In our framework, each user and each item are both represented by the combination of an item embedding and a tag embedding at first. Specifically, item embeddings are learned by Metapath2Vec which is a graph embedding model qualified to embedding heterogeneous information networks. Tag embeddings are learned by a Skip-gram model similar to word embedding. We regarded these embeddings as knowledge embeddings because they both indicate knowledge about the latent relationships of movie-movie and user-movie. At last, a target user's representation and a candidate movie's representation are both fed into a multi-layer perceptron to output the probability that the user likes the item. The probability can be further used to achieve top-n recommendation. The extensive experiments on a movie recommendation dataset demonstrate our framework's superiority over some state-of-the-art recommendation models, especially in the scenario of sparse user-movie interactions.
{"title":"Knowledge Embedding towards the Recommendation with Sparse User-Item Interactions","authors":"Deqing Yang, Ziyi Wang, Junyan Jiang, Yanghua Xiao","doi":"10.1145/3341161.3342876","DOIUrl":"https://doi.org/10.1145/3341161.3342876","url":null,"abstract":"Recently, many researchers in recommender systems have realized that encoding user-item interactions based on deep neural networks (DNNs) promotes collaborative-filtering (CF)'s performance. Nonetheless, those DNN-based models' performance is still limited when observed user-item interactions are very less because the training samples distilled from these interactions are critical for deep learning models. To address this problem, we resort to plenty features distilled from knowledge graphs (KGs), to profile users and items precisely and sufficiently rather than observed user-item interactions. In this paper, we propose a knowledge embedding based recommendation framework to alleviate the problem of sparse user-item interactions in recommendation. In our framework, each user and each item are both represented by the combination of an item embedding and a tag embedding at first. Specifically, item embeddings are learned by Metapath2Vec which is a graph embedding model qualified to embedding heterogeneous information networks. Tag embeddings are learned by a Skip-gram model similar to word embedding. We regarded these embeddings as knowledge embeddings because they both indicate knowledge about the latent relationships of movie-movie and user-movie. At last, a target user's representation and a candidate movie's representation are both fed into a multi-layer perceptron to output the probability that the user likes the item. The probability can be further used to achieve top-n recommendation. The extensive experiments on a movie recommendation dataset demonstrate our framework's superiority over some state-of-the-art recommendation models, especially in the scenario of sparse user-movie interactions.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"136 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115893116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Profile images play an important role in partner selection in a matrimony or dating site. The hypothesis of this paper is that perceived beauty of a profile image is a subjective opinion based on who is viewing the image. We validate this hypothesis by showing that this subjective bias for attractiveness can be learnt from the sender-receiver image pairs. We train a Bi-channel CNN based deep architecture that incorporates the visual features of both users and learns the attractiveness of sender as perceived by the receiver. This network was trained and tested on 3.5 million image pairs and achieved an accuracy of 69% with images alone, thus proving that rather than the eye, beauty lies in the face of the beholder. When this network was used in conjunction with other profile features such as age, city and caste, it further improved the accuracy of the system by a 5% relative number.
{"title":"Beauty lies in the face of the beholder: A Bi-channel CNN architecture for attractiveness modeling in matrimony","authors":"A. Saw, Nitendra Rajput","doi":"10.1145/3341161.3345307","DOIUrl":"https://doi.org/10.1145/3341161.3345307","url":null,"abstract":"Profile images play an important role in partner selection in a matrimony or dating site. The hypothesis of this paper is that perceived beauty of a profile image is a subjective opinion based on who is viewing the image. We validate this hypothesis by showing that this subjective bias for attractiveness can be learnt from the sender-receiver image pairs. We train a Bi-channel CNN based deep architecture that incorporates the visual features of both users and learns the attractiveness of sender as perceived by the receiver. This network was trained and tested on 3.5 million image pairs and achieved an accuracy of 69% with images alone, thus proving that rather than the eye, beauty lies in the face of the beholder. When this network was used in conjunction with other profile features such as age, city and caste, it further improved the accuracy of the system by a 5% relative number.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126383066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pornography can be distributed in multiple forms on the Internet. Online pornography forms a non-negligible fraction of the total Internet traffic, with adult video streaming gaining significant traction among the most visited global websites. Similar to the rise of User Generated Content (UGC) on general Web 2.0 services, adult video service providers have also promoted social interaction and UGC in what is called ‘Porn 2.0'. Discovering the characteristics of Porn 2.0 allows for better understanding of both Internet traffic in general and specifically UGC services. In this paper, using trace-driven analysis, we examined the characteristics of one of the most well-known Porn 2.0 service providers, XHamster. We found that a large proportion of the currently available videos were uploaded in recent years and this has coincided with a rapid growth in the use of video categories. Compared to non-adult UGC services, we found user interaction on XHamster to revolve more strongly around ratings than comments and the average duration and views per video were higher.
{"title":"Measurement and Analysis of an Adult Video Streaming Service","authors":"Yo-Der Song, Mingwei Gong, Aniket Mahanti","doi":"10.1145/3341161.3342940","DOIUrl":"https://doi.org/10.1145/3341161.3342940","url":null,"abstract":"Pornography can be distributed in multiple forms on the Internet. Online pornography forms a non-negligible fraction of the total Internet traffic, with adult video streaming gaining significant traction among the most visited global websites. Similar to the rise of User Generated Content (UGC) on general Web 2.0 services, adult video service providers have also promoted social interaction and UGC in what is called ‘Porn 2.0'. Discovering the characteristics of Porn 2.0 allows for better understanding of both Internet traffic in general and specifically UGC services. In this paper, using trace-driven analysis, we examined the characteristics of one of the most well-known Porn 2.0 service providers, XHamster. We found that a large proportion of the currently available videos were uploaded in recent years and this has coincided with a rapid growth in the use of video categories. Compared to non-adult UGC services, we found user interaction on XHamster to revolve more strongly around ratings than comments and the average duration and views per video were higher.","PeriodicalId":403360,"journal":{"name":"2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126647197","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：