Figure 2 - uploaded by Andrea Ferracani
Content may be subject to copyright.
Example of the user interaction and annotation using a DBPedia resource. The terms highlighted in blues are Wikipedia resources and Facebook users manually annotated.
Source publication
This paper presents a system for the social annotation and discovery of videos based on social networks and social knowledge. The system, developed as a web application, allows users to comment and annotate, manually and automatically, video frames and scenes enriching their content with tags, references to Facebook users and pages and Wikipedia re...
Contexts in source publication
Context 1
... navigation by dragging horizontally the bar on which are shown the individual frames containing annotated comments. The timeline has two levels of precision to scroll through the video respectively every second or every five seconds. Each frame presents two icons of different color that indicate the number of annotations from Facebook and Wikipedia retrieved in the thread of comments. An example of the video player interface is shown in Fig. 3. The recommendation system of videos is based on the analysis of four types of annotations: i) information extracted directly from the user Facebook profile through the Open Graph API, considering in particular the interests and “likes” of pages (this type of information is particularly helpful for suggesting interesting videos during the first access to the system); ii) links to Wikipedia or Facebook pages inserted manually by users in their comments; iii) manual annotations of other users, that belong to the same categories of interest; iv) named entities and topic keywords extracted automatically from text analysis. The manual and automatic annotations are analyzed and categorized by the system using DBPedia ontology struc- ture, in order to propose to users topics and video of interest. Fig. 4 shows an example of semantic user profile automatically generated by the system. Annotations and categories are saved as RDF triples. Profile interests and other informations about network resources are modeled using SPARQL queries. This allows users to identify unex- pected or unknown associations between topics and videos, and gives the possibility to interact with other people who share the same interests. Furthermore users can choose to keep or remove from their semantic profile interests, topics and videos proposed automatically by the system. Finally, the application provides notification systems typical of modern social networks: the user is notified when he is tagged or any of his friends is tagged in a video, or if anyone has tagged one of his videos or videos relating to his interests. Each notification has a visual reference to the frame in the video where the annotation was added. Such actions are also automatically shared on the user’s Facebook Wall. We demonstrate the annotation modalities of the system (Fig. 2), along with its functions to browse and discover new content through the annotations and the implicit suggestions generated by users activity in the network. The main tool for manual annotations is the video annotation widget that provides a simple way to annotate videos (Fig. 3) by users that do not have any knowledge of ontologies, taxonomies or controlled vocabularies, but nevertheless exploits the structured knowledge available in DBPedia or in the Facebook graph through the real-time population of autosuggest input fields, obtained either asynchronously querying the DBPedia endpoint with SPARQL or making REST calls to the Facebook Open Graph API. We also show how entities and topics are extracted automatically when users enter a comment on a frame or respond to a comment in a thread by sending the text to a Java servlet that performs text analysis. This extraction allows the creation of several folksonomies and creates a personalized semantic profile for each user, based on his own interests: this semantic interests profile provides the possibility to watch new videos, access Wikipedia pages or contact Facebook users (Fig. 4). The social aspect of the system stemming from the network of Facebook friends of each user is also considered: it provides suggestions to check new videos either from the annotations of concepts that are part of the interests of each user or by explicit references in the annotations. The demo will also point out how the automatically generated page of each resource in the network (people, videos, named entities, topics) is expressed in RDFa syntax using microformats to bind data to structured vocabularies recog- nized on the Web like DBpedia, the Open Graph Protocol, FOAF 4 and Dublin Core 5 . The focus of the demo of the system is to annotate videos related to entertainment, music and arts but in principle it can be used in other categories such as sports, cars and races. A screencast showing an example of these functionalities is publicly available at: In this demo we have presented a system that allows social network users to discover new videos whose content matches their interest profile. These profiles are automatically created through the semantic analysis of the annotations cre- 4 5 ated by users themselves. Our future work will deal with improved text analysis of user comments, the use of part of speech analysis in order to expand the annotation based on DBPedia, and with exploitation of unsupervised automatic video annotation techniques based on user generated ...
Context 2
... Videos are then transcoded to Flash Video Format on the server side using FFMpeg. The PHP-based Xoom- Stream server has been used to serve videos, without requir- ing the deployment of a full video streaming server. Video thumbnails are created in correspondence with user annotations using FFMpeg. Annotations are stored in two MySQL databases mapping the RDF triples using ARC2 RDF li- brary for PHP 2 . Users can tag resources within comments from both Facebook and Wikipedia using the so-called status tagging fea- ture: typing the @ character in the comment input field they can obtain the list of their friends in the social graph, whilst entering # users can retrieve, using the DBPedia API, a list of Wikipedia pages whose name matches the typed characters. Fig. 2 shows an example of annotation using a DBPedia resource. To reduce the tagging effort required to a user and enrich the semantics, the system performs text analysis to identify potentially interesting tags. Named entities detection is based on the GATE/Annie system 3 and recognizes persons, organizations, places and dates. User annotations are also processed with LDA to identify topics. All these keywords are used to query DBPedia to provide the annotations with links to Wikipedia pages and categories. Visualization and frame-accurate annotation of videos are facilitated by a timeline jQuery widget which allows ...
Similar publications
Background:
It is by no accident The Society’s Psychology of Women Section celebrates thirty-years with a programme which is not only a reflection of “good science”, but of innovative thinking, social-inclusivity, and such diversity. The research seen each year at our Annual Conference, is a reflection of us, the Psychology of Women Section members...
Advertising keywords recommendation is an indispensable component for online advertising with the keywords selected from the target Web pages used for contextual advertising or sponsored search. Several ranking-based algorithms have been proposed for recommending advertising keywords. However, for most of them performance is still lacking, especial...
The paper compares the classical production of printed encyclopedias with the functions and features of their online counterpart Wikipedia. Therefore, it firstly goes into the historic dimension of the encyclopedic principle. On that basis, organisation and work processes of an editorial department are described. In doing so, it secondly highlights...
Web 2.0 has ubiquitously penetrated academia. The dissemination of online information services in higher education has led to substantial changes in faculty teaching methods as well as the learning and study behavior of students. For example, the use of online services, such as Google and Wikipedia, has become mandatory not only during teaching and...
Citations
... Fernadi et al. [33] proposes a deep learning approach that extracts and fuses information across different modalities (status updates, page 'likes', images, relations) for inferring age, gender and personality traits of social media users. Users' annotations and tagging from Wikipedia have been used [34] to model user preferences and suggest videos, users and other resources on a Social Network. Content-based profiling, combined with collaborative filtering, is exploited by the same authors for improving recommendations systems through an hybrid approach in Bertini et al. [22] , Ferracani et al. [35] , 36 ]. ...
In this work we present eSERVANT (eventS in crowdEd places: a smaRt serVice mANagemenT), an ICT hardware and software infrastructure that allows the management of services tied to big events such as concerts, sport matches or fairs that take place in dedicated facilities. Among the main features of eSERVANT there is a vision-based dynamic routing that exploits computer vision techniques to detect people and track their movements within a facility. Thanks to an estimate of room occupancies and inbound/outbound flows, we propose an algorithm for dynamically selecting the best indoor route and avoid hazardous situations. At the same time, we perform user profiling finalized to promoting social networking through recommendation. To assess the usefulness of our infrastructure we established a test case over a six month experimentation and quantitatively evaluated user satisfaction.
... After rejecting his appeal, the Second Circuit Court contended that "so long as the threat on its face and in the circumstances in which it is made is so unequivocal, unconditional, and immediate and specific as to the person threatened, as to concern a gravity of purpose and imminent prospect of execution, the statute may property be applied." 58 Obviously, the explicitly and imminence requirements are paramount. Even much more lenient standard was applied in Planned Parenthood. ...
... Therefore, signature-based multimedia information retrieval is able to attain more robustness in circumstances where authors of a video may intentionally or unintentionally change the video, audio, or text channels of a video message to bypass automatic duplicate copy detection mechanisms than general-purpose information retrieval approaches based on the examination of every bits and bytes of a concerned file. 55 The above approach can be further augmented with crowd-tagging efforts 56,57,58,59 for acquisition of human labels where volunteers all over the world can manually contribute annotations and tags to label any information element of their interests or concerns, whether the element is a picture, audio or video clip, in terms of its relevance to religious extreme topics, events, and activities. A good model that has attained massive success in reality is the practice of Wikipedia where scholars and other professional or amateur communities of interest across the global are voluntarily engaged in a mass crowd-editing effort internationally to develop a major resource for the benefit of the entire human population. ...
This chapter discusses how a variety of software technologies can be leveraged tackle extreme speech on social media. The chapter first overviews Google’s strategy to direct extremist searches towards anti-radicalization websites and comments on its efficacy. The chapter then proceeds to propose two promising technical approaches that can empower private companies and relevant federal agencies to reduce religious extremists’ presence on and impact through social media in a systematic and economically viable way. The first approach leverages content analysis and multimedia mining algorithms to automatically detect content produced by extremists. Using the approach, immediately after a user uploads a document, picture, or video onto a social media platform, a backend engine equipped with a series of state-of-the-art computational content analysis algorithms will be deployed to analyze the types of topics latent in the uploaded material, which will enable social media platforms to discover materials intentionally mislabeled and mis-described by extremists in their attempt to circumvent the traditional keyword or text matching based detection mechanism. Leveraging the same backend content filtering engine, social media platforms can also automatically aggregate multiple files separately updated by one or multiple users independently according to the content similarity between these files. By detecting duplicate or nearly duplicate content on social media through the aforementioned algorithmic engine for automatic content analysis, computers can comprehensively and systematically aggregate otherwise isolated user behaviors associated with individual copies of essentially the same material to produce a consolidated view of content request and consumption for more effective content screening and surveillance. The second approach algorithmically examines collective content consumption behaviors on social media to detect documents, pictures, and videos posted by religious extremists. This approach will detect content posted by religious extremists by observing and analyzing the information propagation pathways and consumption patterns on social media through an automatic and algorithmic approach. Compared with the current self-reporting-based practice, the computational approach can react more efficiently to detect content of concern even when the targeted audience chooses not to cooperate with the social media platform. Additionally, this approach can also save a significant amount of expert labeling effort to train the automatic algorithmic content detector. Overall, the approach will be able to automatically determine the true nature of the content carried by the concerned video regardless of any deceitful labels, titles, or description text the video’s producer or distributor may purposefully associate the video with. The new approach will be able to automatically and systematically detect all videos carrying religiously inciting content in a way that is also much more efficient and comprehensive than the current user self-reporting-based practice. Lastly, the chapter concludes by briefly overviewing three categories of promising software technologies, including natural language processing, machine learning, and social network analysis technologies, which can be tremendously powerful for tackling religious extreme speech on social media.
... After rejecting his appeal, the Second Circuit Court contended that "so long as the threat on its face and in the circumstances in which it is made is so unequivocal, unconditional, and immediate and specific as to the person threatened, as to concern a gravity of purpose and imminent prospect of execution, the statute may property be applied." 58 Obviously, the explicitly and imminence requirements are paramount. Even much more lenient standard was applied in Planned Parenthood. ...
... Therefore, signature-based multimedia information retrieval is able to attain more robustness in circumstances where authors of a video may intentionally or unintentionally change the video, audio, or text channels of a video message to bypass automatic duplicate copy detection mechanisms than general-purpose information retrieval approaches based on the examination of every bits and bytes of a concerned file. 55 The above approach can be further augmented with crowd-tagging efforts 56,57,58,59 for acquisition of human labels where volunteers all over the world can manually contribute annotations and tags to label any information element of their interests or concerns, whether the element is a picture, audio or video clip, in terms of its relevance to religious extreme topics, events, and activities. A good model that has attained massive success in reality is the practice of Wikipedia where scholars and other professional or amateur communities of interest across the global are voluntarily engaged in a mass crowd-editing effort internationally to develop a major resource for the benefit of the entire human population. ...
This chapter examines the viability of free speech norms in the United States and Europe in cubing the surge of extreme speech online. It concludes that, in the United States, criminalizing pro-Jihad propaganda materials that “merely” advocate for the use of violence is likely to be deemed “unconstitutional”. The case law of the European Court on Human Rights, by contrast, permits advocacy for Sharia law and for the establishment of a caliphate, provided the speaker does not call for violence nor does he allude to individuals by names to be attacked. Glorifying Jihad or advocating for Sharia law to advance the interest of any group is strictly prohibited.
... After rejecting his appeal, the Second Circuit Court contended that "so long as the threat on its face and in the circumstances in which it is made is so unequivocal, unconditional, and immediate and specific as to the person threatened, as to concern a gravity of purpose and imminent prospect of execution, the statute may property be applied." 58 Obviously, the explicitly and imminence requirements are paramount. Even much more lenient standard was applied in Planned Parenthood. ...
... Therefore, signature-based multimedia information retrieval is able to attain more robustness in circumstances where authors of a video may intentionally or unintentionally change the video, audio, or text channels of a video message to bypass automatic duplicate copy detection mechanisms than general-purpose information retrieval approaches based on the examination of every bits and bytes of a concerned file. 55 The above approach can be further augmented with crowd-tagging efforts 56,57,58,59 for acquisition of human labels where volunteers all over the world can manually contribute annotations and tags to label any information element of their interests or concerns, whether the element is a picture, audio or video clip, in terms of its relevance to religious extreme topics, events, and activities. A good model that has attained massive success in reality is the practice of Wikipedia where scholars and other professional or amateur communities of interest across the global are voluntarily engaged in a mass crowd-editing effort internationally to develop a major resource for the benefit of the entire human population. ...
This chapter examines the streams of thought contributing to the emergence of the Jihadi Salafism. It highlights the scholarly works of key Salafi scholars who have shaped the ideology of the al-Salafiyya al-Jihadiyya movement as we know it. While the primary sources of Sharia law provide written codes explaining a number of issues capable of being brought before a court, this chapter discusses how the final interpretation of key issues is largely determined by the presiding cleric (sheikh). The scholarly work of those radical clerics effectively enables extremist groups to ideologically justify the use of violent means to achieve their “political” ambitions. Treatment of civilians is a notable example where radical scholars have effectively re-written their own ideologically based rules of war to justify atrocities against innocent population.
... After rejecting his appeal, the Second Circuit Court contended that "so long as the threat on its face and in the circumstances in which it is made is so unequivocal, unconditional, and immediate and specific as to the person threatened, as to concern a gravity of purpose and imminent prospect of execution, the statute may property be applied." 58 Obviously, the explicitly and imminence requirements are paramount. Even much more lenient standard was applied in Planned Parenthood. ...
... Therefore, signature-based multimedia information retrieval is able to attain more robustness in circumstances where authors of a video may intentionally or unintentionally change the video, audio, or text channels of a video message to bypass automatic duplicate copy detection mechanisms than general-purpose information retrieval approaches based on the examination of every bits and bytes of a concerned file. 55 The above approach can be further augmented with crowd-tagging efforts 56,57,58,59 for acquisition of human labels where volunteers all over the world can manually contribute annotations and tags to label any information element of their interests or concerns, whether the element is a picture, audio or video clip, in terms of its relevance to religious extreme topics, events, and activities. A good model that has attained massive success in reality is the practice of Wikipedia where scholars and other professional or amateur communities of interest across the global are voluntarily engaged in a mass crowd-editing effort internationally to develop a major resource for the benefit of the entire human population. ...
The qualitative research in this chapter serves as a supplement and support of the theoretical discourse analysis. It provides a dataset of the most prominent religious clerics in the Middle East and their influence on religious extremist; identifying the names of a hundred and thirty six (136) of the most active Salafi and Salafi Jihadi clerics on social media. The results indicate that there is hardly any difference between how a Salafi or a Jihadi group conduct their propaganda campaigns. Another noticeable result is that both groups use demographic segmentation to specifically target sub-groups, such as women and kids. This chapter concludes that, apart from few negligible differences, both Salafi and jihadi groups are ideologically aligned.
... After rejecting his appeal, the Second Circuit Court contended that "so long as the threat on its face and in the circumstances in which it is made is so unequivocal, unconditional, and immediate and specific as to the person threatened, as to concern a gravity of purpose and imminent prospect of execution, the statute may property be applied." 58 Obviously, the explicitly and imminence requirements are paramount. Even much more lenient standard was applied in Planned Parenthood. ...
... Therefore, signature-based multimedia information retrieval is able to attain more robustness in circumstances where authors of a video may intentionally or unintentionally change the video, audio, or text channels of a video message to bypass automatic duplicate copy detection mechanisms than general-purpose information retrieval approaches based on the examination of every bits and bytes of a concerned file. 55 The above approach can be further augmented with crowd-tagging efforts 56,57,58,59 for acquisition of human labels where volunteers all over the world can manually contribute annotations and tags to label any information element of their interests or concerns, whether the element is a picture, audio or video clip, in terms of its relevance to religious extreme topics, events, and activities. A good model that has attained massive success in reality is the practice of Wikipedia where scholars and other professional or amateur communities of interest across the global are voluntarily engaged in a mass crowd-editing effort internationally to develop a major resource for the benefit of the entire human population. ...
This book is designed to provide specialists, spectators, and students with a brief and engaging exploration of media usage by radical groups and the laws regulating these grey areas of Jihadi propaganda activities. The authors investigate the use of religion to advance political agendas and the legal challenges involved with balancing regulation with free speech rights. The project also examines the reasons behind the limited success of leading initiatives to curb the surge of online extreme speech, such as Google's "Redirect Method" or the U.S. State Department's campaign called "Think Again." The volume concludes by outlining a number of promising technical approaches that can potently empower tech companies to reduce religious extremist groups' presence and impact on social media. © The Editor(s) (if applicable) and The Author(s) 2018. All Rights Reserved.
... There are several collaborative video annotation frameworks, such as [10], [11], [12], [13] and [14]. However, they are mainly concerned with the indexing, annotating, storing and sharing of video data [15]. Hence, they only provide a basis for implementing simple way for video annotation and do not treat concurrent annotation aspects during the collaborative work. ...
Collaborative video annotation system is groupware system which enables a virtual community of participants to share and annotate the same digital video file from geographically dispersed nodes interconnected via the network. The video annotation process allows participants to browse videos, add, delete or update annotations. However, the existing systems are mainly concerned with the indexing, annotating, storing and sharing of video data. Hence, they only provide a basis for implementing simple way for video annotation and do not treat concurrent annotation aspects during the collaborative work. In this paper, we describe AV-Store an original method that combines both advantages of collaborative video annotation and semantic web technologies, and complies with eventual consistency condition when concurrent annotations are performed. The main idea of this approach is to define a new data type where all concurrent annotations commute. The commutativity aims at assuring the consistency of all replicas if participants perform the same sequence of annotations in different orders.
... Social similarity of videos, expressed as popularity distribution across social circles has been used to improve video recommendation in [10]. In [1], the social network of a user is exploited to create his initial user profile and to suggest him videos based on those tagged by his friends using a tool for temporal video commenting; video tagging is also used to update a list of user's interests that are mapped on semantic categories using DBPedia. ...
The recent explosive growth in the use of social networks has raised the question of how to meet the emerging demand for services that address the interests of the users. In this paper we show how considering homophily in social networks can improve video recommendation, using inferred user profiles and modeling users' interests. We propose a socially-aware framework for video commenting, sharing and interest discovery that combines recommendation algorithms, clustering techniques, tools for video tagging and evaluation of tag semantic relatedness. The system allows to connect to friends, curate a personal profile and get video recommendations through a social network.
... Other examples include singer, drummer, guitar player, audience, stage, and close-up, such as the ones provided by youPinkPop, a video search engine for pop concerts [2]. Annotations can also be used to enrich videos using information coming from social networks [1]. Video exploration tools rely on annotations for recommending content to users. ...
Online video sharing systems, such as YouTube, do not provide users enough support to explore community videos that portray people within their social circle. Such services typically consider each video clip as an isolated object, and not as part of a set of related clips. Even though social networks archive media based on higher-order social relationships, they do not provide support for searching and navigating media content that was captured at a particular event by different people. The contribution of this paper is a web-based interface, and the underlying system infrastructure, that allow for socially-aware exploration of media assets from an event. The system combines the best from both video sharing systems and social networks, allowing users to explore and navigate (fragments of) video clips based on their own personal/social interests. The work resulted, as well, in the identification of key requirements for this novel type of socially-aware interfaces. This paper reports on comparative results from a two-phased study conducted during the last four years. The research effort includes a full implementation and deployment of a system, and a series of experiments and user trials that confirmed that our design decisions enable users to explore and view videos they care about.
... Similar work byAart et al. (2009) determines user interest from the Facebook and other SNS platform (Twitter, YouTube 14) and Last.fm 15) ) from NoTube 16) BeanCounter to aggregate users' activities. In the same manner,Bertini et al. (2012), also exploiting user's social graph and their "Like" obtained from Facebook on the purpose of generating semantic user profile. ...
Social media has become one of the most popular media in web and mobile application. In 2011, social networks and blogs are still the top destination of online users, according to a study from Nielsen Company. In their studies, nearly 4 in 5active users visit social network and blog. Social Networks and Blogs sites rule Americans' Internet time, accounting to 23 percent of time spent online. Facebook is the main social network that the U.S internet users spend time more than the other social network services such as Yahoo, Google, AOL Media Network, Twitter, Linked In and so on. In recent trend, most of the companies promote their products in the Facebook by creating the "Facebook Page" that refers to specific product. The "Like" option allows user to subscribed and received updates their interested on from the page. The film makers which produce a lot of films around the world also take part to market and promote their films by exploiting the advantages of using the "Facebook Page". In addition, a great number of streaming service providers allows users to subscribe their service to watch and enjoy movies and TV program. They can instantly watch movies and TV program over the internet to PCs, Macs and TVs. Netflix alone as the world's leading subscription service have more than 30 million streaming members in the United States, Latin America, the United Kingdom and the Nordics. As the matter of facts, a million of movies and TV program with different of genres are offered to the subscriber. In contrast, users need spend a lot time to find the right movies which are related to their interest genre. Recent years there are many researchers who have been propose a method to improve prediction the rating or preference that would give the most related items such as books, music or movies to the garget user or the group of users that have the same interest in the particular items. One of the most popular methods to build recommendation system is traditional Collaborative Filtering (CF). The method compute the similarity of the target user and other users, which then are cluster in the same interest on items according which items that users have been rated. The method then predicts other items from the same group of users to recommend to a group of users. Moreover, There are many items that need to study for suggesting to users such as books, music, movies, news, videos and so on. However, in this paper we only focus on movie as item to recommend to users. In addition, there are many challenges for CF task. Firstly, the "sparsity problem"; it occurs when user information preference is not enough. The recommendation accuracies result is lower compared to the neighbor who composed with a large amount of ratings. The second problem is "cold-start problem"; it occurs whenever new users or items are added into the system, which each has norating or a few rating. For instance, no personalized predictions can be made for a new user without any ratings on the record. In this research we propose a clustering method according to the users' genre interest extracted from social network service (SNS) and user's movies rating information system to solve the "cold-start problem." Our proposed method will clusters the target user together with the other users by combining the user genre interest and the rating information. It is important to realize a huge amount of interesting and useful user's information from Facebook Graph, we can extract information from the "Facebook Page" which "Like" by them. Moreover, we use the Internet Movie Database(IMDb) as the main dataset. The IMDbis online databases that consist of a large amount of information related to movies, TV programs and including actors. This dataset not only used to provide movie information in our Movie Rating Systems, but also as resources to provide movie genre information which extracted from the "Facebook Page". Formerly, the user must login with their Facebook account to login t.....
