Today, mobility is a key feature in the new generation of Internet, which provides a set of custom services through numerous terminals. Smartphones, for example, are a tendency and almost mandatory to anyone living in an urban and modern context. Most of the developed cities have at least one shopping mall full of mobile devices users. These shopping malls provide a number of stores, and people tend to have difficult in finding what they really need. This paper proposes a solution called RecStore. RecStore is a recommendation model to assist customers in reaching what they consider relevant at malls. The recommendation model considers user activities, 330 stores, 30 users and 3 baseline models. The precision, recall and f-measure improved at rates of 118%, 70% and 88% respectively in comparison to the second best model of each metric. Additionally, a mobile application - called InMap - was implemented based on RecStore.
{"title":"RecStore: Recommending Stores for Shopping Mall Customers","authors":"D. V. D. S. Silva, R. S. D. Silva, F. Durão","doi":"10.1145/3126858.3126888","DOIUrl":"https://doi.org/10.1145/3126858.3126888","url":null,"abstract":"Today, mobility is a key feature in the new generation of Internet, which provides a set of custom services through numerous terminals. Smartphones, for example, are a tendency and almost mandatory to anyone living in an urban and modern context. Most of the developed cities have at least one shopping mall full of mobile devices users. These shopping malls provide a number of stores, and people tend to have difficult in finding what they really need. This paper proposes a solution called RecStore. RecStore is a recommendation model to assist customers in reaching what they consider relevant at malls. The recommendation model considers user activities, 330 stores, 30 users and 3 baseline models. The precision, recall and f-measure improved at rates of 118%, 70% and 88% respectively in comparison to the second best model of each metric. Additionally, a mobile application - called InMap - was implemented based on RecStore.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129358706","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}
Carlos de Salles Soares Neto, Thacyla de Sousa Lima, A. L. B. Damasceno, A. Busson
Learning Objects (LOs) are entities that can be used, reused, or referred during the teaching process. LOs allow students to individualize their learning experience with nonlinear browsing mechanisms and content adaptation. The main goal of this tutorial is to discuss both the pedagogical and technological recommendations involved in the authoring of multimedia LOs.
{"title":"Creating Multimedia Learning Objects","authors":"Carlos de Salles Soares Neto, Thacyla de Sousa Lima, A. L. B. Damasceno, A. Busson","doi":"10.1145/3126858.3131626","DOIUrl":"https://doi.org/10.1145/3126858.3131626","url":null,"abstract":"Learning Objects (LOs) are entities that can be used, reused, or referred during the teaching process. LOs allow students to individualize their learning experience with nonlinear browsing mechanisms and content adaptation. The main goal of this tutorial is to discuss both the pedagogical and technological recommendations involved in the authoring of multimedia LOs.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130868166","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}
In this paper we concentrate on the study of the collaborative practices of enthusiasts that create and share subtitles for third-party videos. Based on preliminary results from interviews with some volunteers, we formalize the subtitles creation and sharing process using a business process management model and compare it with other collaborative and crowdsourcing models. We expect that our initial observations can bring a new understanding of the process and, thus, help in the design of next generation video enriching tools.
{"title":"Investigating the Collaborative Process of Subtitles Creation and Sharing for Videos on the Web","authors":"J. Brito, R. Guimarães, Celso A. S. Santos","doi":"10.1145/3126858.3131592","DOIUrl":"https://doi.org/10.1145/3126858.3131592","url":null,"abstract":"In this paper we concentrate on the study of the collaborative practices of enthusiasts that create and share subtitles for third-party videos. Based on preliminary results from interviews with some volunteers, we formalize the subtitles creation and sharing process using a business process management model and compare it with other collaborative and crowdsourcing models. We expect that our initial observations can bring a new understanding of the process and, thus, help in the design of next generation video enriching tools.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123255042","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}
C. V. Araujo, Rayol Mendonca-Neto, F. Nakamura, E. Nakamura
In this paper, we aim at determining whether or not we can predict the success of a music album, based on the comments posted on social networks during 30 days before the album release. For that matter, we focused on the Twitter network for gathering the user comments. As success measures, we considered the Spotify Popularity and the Billboard Units. The reason for those choices is that Spotify represents the most popular type of music consumption today (audio streaming), while Billboard ranking still favors the old school market (physical albums). As a result, we found out that the amount of Positive Tweets (30 days before the album release) can explain 95.5% of the variation in the Spotify Popularity with a simple linear model. On the other hand, we could not find statistical evidence that the volume of comments on Twitter correlates with the album success measured by the Billboard magazine.
{"title":"Predicting Music Success Based on Users' Comments on Online Social Networks","authors":"C. V. Araujo, Rayol Mendonca-Neto, F. Nakamura, E. Nakamura","doi":"10.1145/3126858.3126885","DOIUrl":"https://doi.org/10.1145/3126858.3126885","url":null,"abstract":"In this paper, we aim at determining whether or not we can predict the success of a music album, based on the comments posted on social networks during 30 days before the album release. For that matter, we focused on the Twitter network for gathering the user comments. As success measures, we considered the Spotify Popularity and the Billboard Units. The reason for those choices is that Spotify represents the most popular type of music consumption today (audio streaming), while Billboard ranking still favors the old school market (physical albums). As a result, we found out that the amount of Positive Tweets (30 days before the album release) can explain 95.5% of the variation in the Spotify Popularity with a simple linear model. On the other hand, we could not find statistical evidence that the volume of comments on Twitter correlates with the album success measured by the Billboard magazine.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123418883","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}
Mateus Melo, J. Goebel, Daniel Farias, Cristiano Santos, Tatiana Tavares, G. Corrêa, B. Zatt, M. Porto
This paper discusses results from a quality evaluation experiment involving videos on mobile devices encoded with different configurations of H.264/AVC. The impact of not employing the Fractional Motion Estimation (FME) and the Deblocking Filter (DBF) during the encoding process was also analyzed in the experiments presented in this paper. In order to perform the quality assessment, the objective metrics Structural Similarity (SSIM) and Peak Signal-to-Noise Ratio (PSNR) were used. The subjective evaluation was conducted in two different mobile devices by employing the Mean Opinion Score (MOS) with single stimulus. The obtained results have shown different levels of quality degradation for both modifications. In addition, they led to the conclusion that larger screen devices present a more accentuated drop in subjective quality than small screen devices.
{"title":"Objective and Subjective Video Quality Assessment in Mobile Devices for Low-Complexity H.264/AVC Codecs","authors":"Mateus Melo, J. Goebel, Daniel Farias, Cristiano Santos, Tatiana Tavares, G. Corrêa, B. Zatt, M. Porto","doi":"10.1145/3126858.3131596","DOIUrl":"https://doi.org/10.1145/3126858.3131596","url":null,"abstract":"This paper discusses results from a quality evaluation experiment involving videos on mobile devices encoded with different configurations of H.264/AVC. The impact of not employing the Fractional Motion Estimation (FME) and the Deblocking Filter (DBF) during the encoding process was also analyzed in the experiments presented in this paper. In order to perform the quality assessment, the objective metrics Structural Similarity (SSIM) and Peak Signal-to-Noise Ratio (PSNR) were used. The subjective evaluation was conducted in two different mobile devices by employing the Mean Opinion Score (MOS) with single stimulus. The obtained results have shown different levels of quality degradation for both modifications. In addition, they led to the conclusion that larger screen devices present a more accentuated drop in subjective quality than small screen devices.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128851724","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}
Digital media have been changing fundamentally our society, as a consequence of easier access to contents as well as better and cheaper generation and dissemination through the internet, as witnessed by services such as online videos, games and social networks. More recently, there has been an increasing availability of "smart" services that, among other tasks, help users to locate, understand and analyze automatically media of interest. Smart services are often based on algorithms from data mining and related areas such as machine learning and artificial intelligence. Beyond the efficiency and effectiveness of theses services, there is a growing concern about the fairness, accountability and transparency associated with them, which is the subject of this talk. Fairness comprises guarantees that algorithms are neither biased nor discriminatory, even when they are mathematically and computationally correct. Accountability means the identification of entities, human or not, that should be held responsible for the algorithms' consequences. Transparency is the property of generating understandable explanations on the algorithms' outcomes. In this talk we are going to discuss and characterize data mining algorithms, in particular when applied to web and social networks, with respect to fairness, accountability and transparency, and present strategies that assure these properties while satisfying other usual requirements such as precision, effectiveness, and privacy preservation.
{"title":"Fairness, Accountability, and Transparency while Mining Data from the Web and Social Networks","authors":"Wagner Meira Jr","doi":"10.1145/3126858.3133314","DOIUrl":"https://doi.org/10.1145/3126858.3133314","url":null,"abstract":"Digital media have been changing fundamentally our society, as a consequence of easier access to contents as well as better and cheaper generation and dissemination through the internet, as witnessed by services such as online videos, games and social networks. More recently, there has been an increasing availability of \"smart\" services that, among other tasks, help users to locate, understand and analyze automatically media of interest. Smart services are often based on algorithms from data mining and related areas such as machine learning and artificial intelligence. Beyond the efficiency and effectiveness of theses services, there is a growing concern about the fairness, accountability and transparency associated with them, which is the subject of this talk. Fairness comprises guarantees that algorithms are neither biased nor discriminatory, even when they are mathematically and computationally correct. Accountability means the identification of entities, human or not, that should be held responsible for the algorithms' consequences. Transparency is the property of generating understandable explanations on the algorithms' outcomes. In this talk we are going to discuss and characterize data mining algorithms, in particular when applied to web and social networks, with respect to fairness, accountability and transparency, and present strategies that assure these properties while satisfying other usual requirements such as precision, effectiveness, and privacy preservation.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115732014","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}
Kamila R. H. Rodrigues, C. C. Viel, Isabela Zaine, Bruna C. R. Cunha, L. Scalco, M. G. Pimentel
Mobile computing can be a facilitator for collecting data due to the fact that users carry their smartphones almost everywhere and all the time and because they can collect a wide range of data -- textual, audiovisual and data collected automatically by sensors. Considering this opportunity, we developed the ESPIM (Experience Sampling and Programmed Intervention Method), an computer-aided method for programming multimedia data collection forms and carry out remote interventions. Using the ESPIM, professionals of areas such as healthcare and education can plan data collection and define intervention programs using methods and procedures from their own areas. The programs containing the queries and tasks are retrieved by a mobile application installed in the devices of users who participate in the data collection. The mobile application runs the programs according to queries and tasks planned by the specialists. Both queries and responses can contain text, audio, and video data. In this paper we discuss about the technological infrastructure used in ESPIM system and also about the preliminary results obtained through tests and evaluation carried out with stakeholders of the target population. These results allowed us carried out improvements in the system.
{"title":"Data Collection and Intervention Personalized as Interactive Multimedia Documents","authors":"Kamila R. H. Rodrigues, C. C. Viel, Isabela Zaine, Bruna C. R. Cunha, L. Scalco, M. G. Pimentel","doi":"10.1145/3126858.3131574","DOIUrl":"https://doi.org/10.1145/3126858.3131574","url":null,"abstract":"Mobile computing can be a facilitator for collecting data due to the fact that users carry their smartphones almost everywhere and all the time and because they can collect a wide range of data -- textual, audiovisual and data collected automatically by sensors. Considering this opportunity, we developed the ESPIM (Experience Sampling and Programmed Intervention Method), an computer-aided method for programming multimedia data collection forms and carry out remote interventions. Using the ESPIM, professionals of areas such as healthcare and education can plan data collection and define intervention programs using methods and procedures from their own areas. The programs containing the queries and tasks are retrieved by a mobile application installed in the devices of users who participate in the data collection. The mobile application runs the programs according to queries and tasks planned by the specialists. Both queries and responses can contain text, audio, and video data. In this paper we discuss about the technological infrastructure used in ESPIM system and also about the preliminary results obtained through tests and evaluation carried out with stakeholders of the target population. These results allowed us carried out improvements in the system.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121030869","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}
The need to retrieve information in multimedia content increases the demand for systems that use automatic speech recognition. A speech recognition system enables the computer to interpret audio signals, generating approximate textual transcriptions. These systems are based on probabilistic models that create a robust and correct model for human speech. In this paper it is presented a speech recognition systems architecture and a description of its basic components: the acoustic model, language model, lexical and decoder. The training process of acoustic and language models is also presented. Finally, it its presented how these systems can be used in several applications.
{"title":"Use of Automatic Speech Recognition Systems for Multimedia Applications","authors":"Marcos Valadão Gualberto Ferreira, J. Souza","doi":"10.1145/3126858.3131630","DOIUrl":"https://doi.org/10.1145/3126858.3131630","url":null,"abstract":"The need to retrieve information in multimedia content increases the demand for systems that use automatic speech recognition. A speech recognition system enables the computer to interpret audio signals, generating approximate textual transcriptions. These systems are based on probabilistic models that create a robust and correct model for human speech. In this paper it is presented a speech recognition systems architecture and a description of its basic components: the acoustic model, language model, lexical and decoder. The training process of acoustic and language models is also presented. Finally, it its presented how these systems can be used in several applications.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115744006","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}
Among the several vehicles of social communication, digital signage displays have been playing a remarkable role in both public and private spaces. Such Digital Out-of-Home (DOOH) media allows for the rapid dissemination of collective information to a large number of people. It is observed, however, that there is a large distance between the graphical abstractions offered by DOOH authoring tools and the underlying language for the representation of hyper/multimedia content. Document representation becomes complex, sometimes makes use of scripting languages, and therefore is illegible by authors and even difficult for automated information extraction. In this context, this paper proposes STorM, a hypermedia model and its language STorML that defines higher-level entities related to the concepts found in the audiovisual industry, such as scenes, tracks and media.
{"title":"STorM: A Hypermedia Authoring Model for Interactive Digital Out-of-Home Media","authors":"Marco A. Freesz, L. Yung, M. Moreno","doi":"10.1145/3126858.3126889","DOIUrl":"https://doi.org/10.1145/3126858.3126889","url":null,"abstract":"Among the several vehicles of social communication, digital signage displays have been playing a remarkable role in both public and private spaces. Such Digital Out-of-Home (DOOH) media allows for the rapid dissemination of collective information to a large number of people. It is observed, however, that there is a large distance between the graphical abstractions offered by DOOH authoring tools and the underlying language for the representation of hyper/multimedia content. Document representation becomes complex, sometimes makes use of scripting languages, and therefore is illegible by authors and even difficult for automated information extraction. In this context, this paper proposes STorM, a hypermedia model and its language STorML that defines higher-level entities related to the concepts found in the audiovisual industry, such as scenes, tracks and media.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":" 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113952348","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}
Recommender systems are widely used to minimize the information overload problem. A great source of information is users' reviews, since they provide both item descriptions and users' opinions. Recent works that process reviews often neglect problems such as polysemy and sinonimy. On the other hand, systems that rely on word sense disambiguation focus their efforts on items's static descriptions. In this paper, we propose a hybrid recommender system that uses word sense disambiguation and entity linking to produce concept-based item representations extracted from users' reviews. Our findings suggest that adding such semantics to items' representations have a positive impact on recommendations.
{"title":"Semantic Organization of User's Reviews Applied in Recommender Systems","authors":"Ronnie S. Marinho, R. M. D'Addio, M. Manzato","doi":"10.1145/3126858.3131600","DOIUrl":"https://doi.org/10.1145/3126858.3131600","url":null,"abstract":"Recommender systems are widely used to minimize the information overload problem. A great source of information is users' reviews, since they provide both item descriptions and users' opinions. Recent works that process reviews often neglect problems such as polysemy and sinonimy. On the other hand, systems that rely on word sense disambiguation focus their efforts on items's static descriptions. In this paper, we propose a hybrid recommender system that uses word sense disambiguation and entity linking to produce concept-based item representations extracted from users' reviews. Our findings suggest that adding such semantics to items' representations have a positive impact on recommendations.","PeriodicalId":338362,"journal":{"name":"Proceedings of the 23rd Brazillian Symposium on Multimedia and the Web","volume":"519 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134315404","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：