Inference of online social network users' attributes and interests has been an active research to... more Inference of online social network users' attributes and interests has been an active research topic. Accurate identification of users' attributes and interests is crucial for improving the performance of personalization and recommender systems. Most of the existing works have focused on textual content generated by the users and have successfully used it for predicting users' interests and other identifying attributes. However, little attention has been paid to user generated visual content (images) that is becoming increasingly popular and pervasive in recent times. We posit that images posted by users on online social networks are a reflection of topics they are interested in and propose an approach to infer user attributes from images posted by them. We analyze the content of individual images and then aggregate the image-level knowledge to infer user-level interest distribution. We employ image-level similarity to propagate the label information between images, as well as utilize the image category information derived from the user created organization structure to further propagate the category-level knowledge for all images. A real life social network dataset created from Pinterest is used for evaluation and the experimental results demonstrate the effectiveness of our proposed approach.
Proceedings of the Thirteenth International Workshop, Aug 11, 2013
ABSTRACT Online social networks have attracted attention of people from both the academia and rea... more ABSTRACT Online social networks have attracted attention of people from both the academia and real world. In particular, the rich multimedia information accumulated in recent years provides an easy and convenient way for more active communication between people. This offers an opportunity to research people's behaviors and activities based on those multimedia content, which can be considered as social imagematics. One emerging area is driven by the fact that these massive multimedia data contain people's daily sentiments and opinions. However, existing sentiment analysis typically only pays attention to the textual information regardless of the visual content, which may be more informative in expressing people's sentiments and opinions. In this paper, we attempt to analyze the online sentiment changes of social media users using both the textual and visual content. In particular, we analyze the sentiment changes of Twitter users using both textual and visual features. An empirical study of real Twitter data sets indicates that the sentiments expressed in textual content and visual content are correlated. The preliminary results in this paper give insight into the important role of visual content in online social media.
Frontiers of Computer Science in China, Apr 25, 2011
The construction process for a similarity matrix has an important impact on the performance of sp... more The construction process for a similarity matrix has an important impact on the performance of spectral clustering algorithms. In this paper, we propose a random walk based approach to process the Gaussian kernel similarity matrix. In this method, the pair-wise similarity between two data points is not only related to the two points, but also related to their neighbors. As a result, the new similarity matrix is closer to the ideal matrix which can provide the best clustering result. We give a theoretical analysis of the similarity matrix and apply this similarity matrix to spectral clustering. We also propose a method to handle noisy items which may cause deterioration of clustering performance. Experimental results on real-world data sets show that the proposed spectral clustering algorithm significantly outperforms existing algorithms.
Inference of online social network users' attributes and interests has been an active research to... more Inference of online social network users' attributes and interests has been an active research topic. Accurate identification of users' attributes and interests is crucial for improving the performance of personalization and recommender systems. Most of the existing works have focused on textual content generated by the users and have successfully used it for predicting users' interests and other identifying attributes. However, little attention has been paid to user generated visual content (images) that is becoming increasingly popular and pervasive in recent times. We posit that images posted by users on online social networks are a reflection of topics they are interested in and propose an approach to infer user attributes from images posted by them. We analyze the content of individual images and then aggregate the image-level knowledge to infer user-level interest distribution. We employ image-level similarity to propagate the label information between images, as well as utilize the image category information derived from the user created organization structure to further propagate the category-level knowledge for all images. A real life social network dataset created from Pinterest is used for evaluation and the experimental results demonstrate the effectiveness of our proposed approach.
The construction process for a similarity matrix has an important impact on the performance of sp... more The construction process for a similarity matrix has an important impact on the performance of spectral clustering algorithms. In this paper, we propose a random walk based approach to process the Gaussian kernel similarity matrix. In this method, the pair-wise similarity between two data points is not only related to the two points, but also related to their neighbors. As a result, the new similarity matrix is closer to the ideal matrix which can provide the best clustering result. We give a theoretical analysis of the similarity matrix and apply this similarity matrix to spectral clustering. We also propose a method to handle noisy items which may cause deterioration of clustering performance. Experimental results on real-world data sets show that the proposed spectral clustering algorithm significantly outperforms existing algorithms.
2011 IEEE 11th International Conference on Data Mining, 2011
To alleviate the memory and computational burdens of spectral clustering for large scale problems... more To alleviate the memory and computational burdens of spectral clustering for large scale problems, some kind of low-rank matrix approximation is usually employed. Nyström method is an efficient technique to generate lowrank matrix approximation and its most important aspect is sampling. The matrix approximation errors of several sampling schemes have been theoretically analyzed for a number of learning tasks. However, the impact of matrix approximation error on the clustering performance of spectral clustering has not been studied. In this paper, we firstly analyze the performance of Nyström method in terms of clusterability, thus answer the impact of matrix approximation error on the clustering performance of spectral clustering. Our analysis immediately suggests an incremental sampling scheme for the Nyström method based spectral clustering. Experimental results show that the proposed incremental sampling scheme outperforms existing sampling schemes on various clustering tasks and image segmentation applications, and its efficiency is comparable with existing sampling schemes.
Proceedings of the Second International Workshop on Issues of Sentiment Discovery and Opinion Mining - WISDOM '13, 2013
ABSTRACT Visual content analysis has always been important yet challenging. Thanks to the popular... more ABSTRACT Visual content analysis has always been important yet challenging. Thanks to the popularity of social networks, images become an convenient carrier for information diffusion among online users. To understand the diffusion patterns and different aspects of the social images, we need to interpret the images first. Similar to textual content, images also carry different levels of sentiment to their viewers. However, different from text, where sentiment analysis can use easily accessible semantic and context information, how to extract and interpret the sentiment of an image remains quite challenging. In this paper, we propose an image sentiment prediction framework, which leverages the mid-level attributes of an image to predict its sentiment. This makes the sentiment classification results more interpretable than directly using the low-level features of an image. To obtain a better performance on images containing faces, we introduce eigenface-based facial expression detection as an additional mid-level attributes. An empirical study of the proposed framework shows improved performance in terms of prediction accuracy. More importantly, by inspecting the prediction results, we are able to discover interesting relationships between mid-level attribute and image sentiment.
ABSTRACT Vehicle routing usually depends on a road map, and road maps are expensive to create and... more ABSTRACT Vehicle routing usually depends on a road map, and road maps are expensive to create and maintain. While crowdsourcing road maps from logged GPS data has proven effective, the limited availability of GPS data limits their coverage area. To overcome this limitation, we show how to use location data from geotagged tweets, which cover much of the world, to compute routes directly without making a road map. We compensate for the wide spacing of tweets' latitude/longitude points by using probabilistic time geography, which explicitly models the uncertain location of someone traveling between measured locations. In our formulation, each pair of temporally adjacent tweets contributes an estimate of the driving time along hypothesised roads in a regular grid. We show how to compute these estimates as expected values based on probabilistic Brownian bridges. We can compute routes on this regular grid using traditional A* search. Our experiments demonstrate that our computed routes match well with routes computed on the actual road network using a commercial router. Furthermore, we show that our computed routes vary sensibly with changes in traffic between rush hour and weekends. We also apply the same technique to compute reasonable airplane routes.
Proceedings of the Thirteenth International Workshop on Multimedia Data Mining - MDMKDD '13, 2013
ABSTRACT Online social networks have attracted attention of people from both the academia and rea... more ABSTRACT Online social networks have attracted attention of people from both the academia and real world. In particular, the rich multimedia information accumulated in recent years provides an easy and convenient way for more active communication between people. This offers an opportunity to research people's behaviors and activities based on those multimedia content, which can be considered as social imagematics. One emerging area is driven by the fact that these massive multimedia data contain people's daily sentiments and opinions. However, existing sentiment analysis typically only pays attention to the textual information regardless of the visual content, which may be more informative in expressing people's sentiments and opinions. In this paper, we attempt to analyze the online sentiment changes of social media users using both the textual and visual content. In particular, we analyze the sentiment changes of Twitter users using both textual and visual features. An empirical study of real Twitter data sets indicates that the sentiments expressed in textual content and visual content are correlated. The preliminary results in this paper give insight into the important role of visual content in online social media.
2013 IEEE 13th International Conference on Data Mining Workshops, 2013
ABSTRACT In recent years, political campaigns have paid increasing attention to social media. Dur... more ABSTRACT In recent years, political campaigns have paid increasing attention to social media. During the election period, numerous election related images are posted. However, not all the images have the same effectiveness, and researchers have not investigated the intrinsic relationship between the effectiveness and the high-level visual features of social images. In this paper, we present a new study to analyze the effectiveness of election related images in social media. We first compute three semantic visual attributes for election related images: 1) face attribute, which indicates the presence of a political candidate, 2) text attribute, which describes the area of text information, 3) logo attribute, which denotes whether an image contains a campaign logo. Next, we consider the effectiveness in terms of the number of views and comments, and employ analysis of variance and association analysis to understand the importance of visual attributes in affecting the effectiveness of election related images. In addition, visual attributes distribution analysis reveals Obama campaign's deliberate effort targeting social media. The experiments on the 2012 US presidential election related images provide interesting insight that can be exploited in similar scenarios.
Proceedings of the 2nd international workshop on Socially-aware multimedia - SAM '13, 2013
ABSTRACT In recent years, online social networks become popular across the world. The success of ... more ABSTRACT In recent years, online social networks become popular across the world. The success of online social networks, such as Twitter and Facebook, have led to the emergence of similar social networks in many countries with different cultural backgrounds. For instance, Sina Weibo and Tencent Weibo are the most popular microblog services in China. One interesting question is what are the similarities and differences between these social networks. In particular, we are interested in finding out whether similar types of information propagates similarly or not on a similar kind of social network service. In this paper, we analyze two representative microblog service providers, Twitter from U.S. and Tencent Weibo from China. We focus on the patterns of event-driven information propagation. We employ several metrics to measure the differences in information propagation patterns on a variety of selected topic categories. Surprisingly, the preliminary results of our study show that there is no significant difference between the two platforms in terms of information propagation patterns. This opens up further investigations to understand the factors as work.
Inference of online social network users' attributes and interests has been an active research to... more Inference of online social network users' attributes and interests has been an active research topic. Accurate identification of users' attributes and interests is crucial for improving the performance of personalization and recommender systems. Most of the existing works have focused on textual content generated by the users and have successfully used it for predicting users' interests and other identifying attributes. However, little attention has been paid to user generated visual content (images) that is becoming increasingly popular and pervasive in recent times. We posit that images posted by users on online social networks are a reflection of topics they are interested in and propose an approach to infer user attributes from images posted by them. We analyze the content of individual images and then aggregate the image-level knowledge to infer user-level interest distribution. We employ image-level similarity to propagate the label information between images, as well as utilize the image category information derived from the user created organization structure to further propagate the category-level knowledge for all images. A real life social network dataset created from Pinterest is used for evaluation and the experimental results demonstrate the effectiveness of our proposed approach.
Proceedings of the Thirteenth International Workshop, Aug 11, 2013
ABSTRACT Online social networks have attracted attention of people from both the academia and rea... more ABSTRACT Online social networks have attracted attention of people from both the academia and real world. In particular, the rich multimedia information accumulated in recent years provides an easy and convenient way for more active communication between people. This offers an opportunity to research people's behaviors and activities based on those multimedia content, which can be considered as social imagematics. One emerging area is driven by the fact that these massive multimedia data contain people's daily sentiments and opinions. However, existing sentiment analysis typically only pays attention to the textual information regardless of the visual content, which may be more informative in expressing people's sentiments and opinions. In this paper, we attempt to analyze the online sentiment changes of social media users using both the textual and visual content. In particular, we analyze the sentiment changes of Twitter users using both textual and visual features. An empirical study of real Twitter data sets indicates that the sentiments expressed in textual content and visual content are correlated. The preliminary results in this paper give insight into the important role of visual content in online social media.
Frontiers of Computer Science in China, Apr 25, 2011
The construction process for a similarity matrix has an important impact on the performance of sp... more The construction process for a similarity matrix has an important impact on the performance of spectral clustering algorithms. In this paper, we propose a random walk based approach to process the Gaussian kernel similarity matrix. In this method, the pair-wise similarity between two data points is not only related to the two points, but also related to their neighbors. As a result, the new similarity matrix is closer to the ideal matrix which can provide the best clustering result. We give a theoretical analysis of the similarity matrix and apply this similarity matrix to spectral clustering. We also propose a method to handle noisy items which may cause deterioration of clustering performance. Experimental results on real-world data sets show that the proposed spectral clustering algorithm significantly outperforms existing algorithms.
Inference of online social network users' attributes and interests has been an active research to... more Inference of online social network users' attributes and interests has been an active research topic. Accurate identification of users' attributes and interests is crucial for improving the performance of personalization and recommender systems. Most of the existing works have focused on textual content generated by the users and have successfully used it for predicting users' interests and other identifying attributes. However, little attention has been paid to user generated visual content (images) that is becoming increasingly popular and pervasive in recent times. We posit that images posted by users on online social networks are a reflection of topics they are interested in and propose an approach to infer user attributes from images posted by them. We analyze the content of individual images and then aggregate the image-level knowledge to infer user-level interest distribution. We employ image-level similarity to propagate the label information between images, as well as utilize the image category information derived from the user created organization structure to further propagate the category-level knowledge for all images. A real life social network dataset created from Pinterest is used for evaluation and the experimental results demonstrate the effectiveness of our proposed approach.
The construction process for a similarity matrix has an important impact on the performance of sp... more The construction process for a similarity matrix has an important impact on the performance of spectral clustering algorithms. In this paper, we propose a random walk based approach to process the Gaussian kernel similarity matrix. In this method, the pair-wise similarity between two data points is not only related to the two points, but also related to their neighbors. As a result, the new similarity matrix is closer to the ideal matrix which can provide the best clustering result. We give a theoretical analysis of the similarity matrix and apply this similarity matrix to spectral clustering. We also propose a method to handle noisy items which may cause deterioration of clustering performance. Experimental results on real-world data sets show that the proposed spectral clustering algorithm significantly outperforms existing algorithms.
2011 IEEE 11th International Conference on Data Mining, 2011
To alleviate the memory and computational burdens of spectral clustering for large scale problems... more To alleviate the memory and computational burdens of spectral clustering for large scale problems, some kind of low-rank matrix approximation is usually employed. Nyström method is an efficient technique to generate lowrank matrix approximation and its most important aspect is sampling. The matrix approximation errors of several sampling schemes have been theoretically analyzed for a number of learning tasks. However, the impact of matrix approximation error on the clustering performance of spectral clustering has not been studied. In this paper, we firstly analyze the performance of Nyström method in terms of clusterability, thus answer the impact of matrix approximation error on the clustering performance of spectral clustering. Our analysis immediately suggests an incremental sampling scheme for the Nyström method based spectral clustering. Experimental results show that the proposed incremental sampling scheme outperforms existing sampling schemes on various clustering tasks and image segmentation applications, and its efficiency is comparable with existing sampling schemes.
Proceedings of the Second International Workshop on Issues of Sentiment Discovery and Opinion Mining - WISDOM '13, 2013
ABSTRACT Visual content analysis has always been important yet challenging. Thanks to the popular... more ABSTRACT Visual content analysis has always been important yet challenging. Thanks to the popularity of social networks, images become an convenient carrier for information diffusion among online users. To understand the diffusion patterns and different aspects of the social images, we need to interpret the images first. Similar to textual content, images also carry different levels of sentiment to their viewers. However, different from text, where sentiment analysis can use easily accessible semantic and context information, how to extract and interpret the sentiment of an image remains quite challenging. In this paper, we propose an image sentiment prediction framework, which leverages the mid-level attributes of an image to predict its sentiment. This makes the sentiment classification results more interpretable than directly using the low-level features of an image. To obtain a better performance on images containing faces, we introduce eigenface-based facial expression detection as an additional mid-level attributes. An empirical study of the proposed framework shows improved performance in terms of prediction accuracy. More importantly, by inspecting the prediction results, we are able to discover interesting relationships between mid-level attribute and image sentiment.
ABSTRACT Vehicle routing usually depends on a road map, and road maps are expensive to create and... more ABSTRACT Vehicle routing usually depends on a road map, and road maps are expensive to create and maintain. While crowdsourcing road maps from logged GPS data has proven effective, the limited availability of GPS data limits their coverage area. To overcome this limitation, we show how to use location data from geotagged tweets, which cover much of the world, to compute routes directly without making a road map. We compensate for the wide spacing of tweets' latitude/longitude points by using probabilistic time geography, which explicitly models the uncertain location of someone traveling between measured locations. In our formulation, each pair of temporally adjacent tweets contributes an estimate of the driving time along hypothesised roads in a regular grid. We show how to compute these estimates as expected values based on probabilistic Brownian bridges. We can compute routes on this regular grid using traditional A* search. Our experiments demonstrate that our computed routes match well with routes computed on the actual road network using a commercial router. Furthermore, we show that our computed routes vary sensibly with changes in traffic between rush hour and weekends. We also apply the same technique to compute reasonable airplane routes.
Proceedings of the Thirteenth International Workshop on Multimedia Data Mining - MDMKDD '13, 2013
ABSTRACT Online social networks have attracted attention of people from both the academia and rea... more ABSTRACT Online social networks have attracted attention of people from both the academia and real world. In particular, the rich multimedia information accumulated in recent years provides an easy and convenient way for more active communication between people. This offers an opportunity to research people's behaviors and activities based on those multimedia content, which can be considered as social imagematics. One emerging area is driven by the fact that these massive multimedia data contain people's daily sentiments and opinions. However, existing sentiment analysis typically only pays attention to the textual information regardless of the visual content, which may be more informative in expressing people's sentiments and opinions. In this paper, we attempt to analyze the online sentiment changes of social media users using both the textual and visual content. In particular, we analyze the sentiment changes of Twitter users using both textual and visual features. An empirical study of real Twitter data sets indicates that the sentiments expressed in textual content and visual content are correlated. The preliminary results in this paper give insight into the important role of visual content in online social media.
2013 IEEE 13th International Conference on Data Mining Workshops, 2013
ABSTRACT In recent years, political campaigns have paid increasing attention to social media. Dur... more ABSTRACT In recent years, political campaigns have paid increasing attention to social media. During the election period, numerous election related images are posted. However, not all the images have the same effectiveness, and researchers have not investigated the intrinsic relationship between the effectiveness and the high-level visual features of social images. In this paper, we present a new study to analyze the effectiveness of election related images in social media. We first compute three semantic visual attributes for election related images: 1) face attribute, which indicates the presence of a political candidate, 2) text attribute, which describes the area of text information, 3) logo attribute, which denotes whether an image contains a campaign logo. Next, we consider the effectiveness in terms of the number of views and comments, and employ analysis of variance and association analysis to understand the importance of visual attributes in affecting the effectiveness of election related images. In addition, visual attributes distribution analysis reveals Obama campaign's deliberate effort targeting social media. The experiments on the 2012 US presidential election related images provide interesting insight that can be exploited in similar scenarios.
Proceedings of the 2nd international workshop on Socially-aware multimedia - SAM '13, 2013
ABSTRACT In recent years, online social networks become popular across the world. The success of ... more ABSTRACT In recent years, online social networks become popular across the world. The success of online social networks, such as Twitter and Facebook, have led to the emergence of similar social networks in many countries with different cultural backgrounds. For instance, Sina Weibo and Tencent Weibo are the most popular microblog services in China. One interesting question is what are the similarities and differences between these social networks. In particular, we are interested in finding out whether similar types of information propagates similarly or not on a similar kind of social network service. In this paper, we analyze two representative microblog service providers, Twitter from U.S. and Tencent Weibo from China. We focus on the patterns of event-driven information propagation. We employ several metrics to measure the differences in information propagation patterns on a variety of selected topic categories. Surprisingly, the preliminary results of our study show that there is no significant difference between the two platforms in terms of information propagation patterns. This opens up further investigations to understand the factors as work.
Uploads
Papers by Quanzeng You