Automatic playlist generation based on tracking user’s listening habits

Algorithms for automatic playlist generation solve the problem of tedious and time consuming manual selection of musical playlists. These algorithms generate playlists according to the user’s music preferences of the moment. The user describes his preferences either by manually inputting a couple of example songs, or by defining constraints for the choice of music. The approaches to automatic playlist generation up to now were based on examining the metadata attached to the music pieces. Some of them took also the listening history into account. But anyway, a heavy accent has been put on the metadata, while the listening history, if it was used at all, had a minor role. Missings and errors in metadata frequently appear, especially when the music is acquired from the Internet. When the metadata is missing or wrong, the approaches proposed so far cannot work. Besides, entering constraints for the playlist generation can be a difficult activity. In our approach we ignored the metadata and focused on examining the listening habits. We developed two simple algorithms that track the listening habits and form a listener model—a profile of listening habits. The listener model is then used for automatic playlist generation. We developed a simple media player which tracks the listening habits and generates playlists according to the listener model. We tried the solution with a group of users. The experiment was not a successful one, but it threw some new light on the relationship between the listening habits and playlist generation.     

A collaborative user-centered framework for recommending items in Online Social Networks

Recommender Systems are more and more playing an important role in our life, representing useful tools helping users to find “what they need” from a very large number of candidates and supporting people in making decisions in various contexts: what items to buy, which movie to watch, or even who they can invite to their social network, etc. In this paper, we propose a novel collaborative user-centered recommendation approach in which several aspects related to users and available in Online Social Networks – i.e. preferences (usually in the shape of items’ metadata), opinions (textual comments to which it is possible to associate a sentiment), behavior (in the majority of cases logs of past items’ observations made by users), feedbacks (usually expressed in the form of ratings) – are considered and integrated together with items’ features and context information within a general framework that can support different applications using proper customizations (e.g., recommendation of news, photos, movies, travels, etc.). Experiments on system accuracy and user satisfaction in several domains shows how our approach provides very promising and interesting results.     

FESORIA: An integrated system for analysis,management and smart presentation of audio/video streaming services

Due to the elevated consumption of resources, the high cost of the production of contents and the quality of service required in audio/video streaming services, it is extremely important to optimize all the elements involved in the deployment of these services. With this goal in mind, provider companies have developed their management and presentation tools. At the same time, some specific tools for audio/video streaming analysis have appeared. Data are collected from servers and proxies by analyzing their log files in order to generate different types of reports. In spite of their utility, there is a disconnection between these types of tools. In this way, several important relationships between collected data are lost and the influence of other important aspects such as the behaviour of the users and their relationship with the subject or the length of the contents is not considered. This generates inaccurate analyses and the impossibility to improve the presentation, for example by generating recommendations using the information gathered from the analysis tool. Fesoria is a system which combines both characteristics. It is an analysis tool and, at the same time, a system to manage the whole audio/video service. Fesoria is able to process the logs gathered from the streaming servers and proxies, and combine the extracted information with other types of data, such as content metadata, content distribution networks architecture, user preferences, etc. All this information is analyzed in order to generate reports on service performance, access evolution and users’ preferences, and thus to improve the presentation of the services. The system has been used in real audio/video services since 2001 with satisfactory results.

A feature-based approach to better automatic treebank conversion

In the field of constituency parsing, there exist multiple human-labeled treebanks which are built on non-overlapping text samples and follow different annotation standards. Due to the extreme cost of annotating parse trees by human, it is desirable to automatically convert one treebank (called source treebank) to the standard of another treebank (called target treebank) which we are interested in. Conversion results can be manually corrected to obtain higher-quality annotations or can be directly used as additional training data for building syntactic parsers. To perform automatic treebank conversion, we divide constituency parses into two separate levels: the part-of-speech (POS) and syntactic structure (bracketing structures and constituent labels), and conduct conversion on these two levels respectively with a feature-based approach. The basic idea of the approach is to encode original annotations in a source treebank as guide features during the conversion process. Experiments on two Chinese treebanks show that our approach can convert POS tags and syntactic structures with the accuracy of 96.6 and 84.8 %, respectively, which are the best reported results on this task.     

Learning with multi-resolution overlapping communities

A recent surge of participatory web and social media has created a new laboratory for studying human relations and collective behavior on an unprecedented scale. In this work, we study the predictive power of social connections to determine the preferences or behaviors of individuals such as whether a user supports a certain political view, whether one likes a product, whether she would like to vote for a presidential candidate, etc. Since an actor is likely to participate in multiple different communities with each regulating the actor’s behavior in varying degrees, and a natural hierarchy might exist between these communities, we propose to zoom into a network at multiple different resolutions and determine which communities reflect a targeted behavior. We develop an efficient algorithm to extract a hierarchy of overlapping communities. Empirical results on social media networks demonstrate the promising potential of the proposed approach in real-world applications.     

Semantic-based framework for personalised ambient media

The paper proposes a semantic-based metadata framework for personalised interaction with TV media in a connected home context. Our approach allows the current home media centres to go beyond the simple concept of electronic programme guides and to offer the users a personalised media experience in an ambient home environment. The user’s characteristics, preferences and context are used to personalise the user’s experience of viewing and interacting with multimedia content on different heterogeneous devices. The TV-Anytime specification provides the basis for the metadata framework for handling content from IP, digital broadcast, and Blu-ray disc sources.     

基于流媒体的元数据加速下载技术的优化研究

为了提高基于流媒体的元数据下载速度,采用一种基于客户/服务器探测变化机制的时间优先多点下载技术,用户可以获得当前连接条件下最优质的流媒体质量,并通过测试验证时间优先多点下载技术能较好地提高元数据的下载速度,保证了流媒体服务质量.     

Incentivized provision of metadata, semantic reasoning and time-driven filtering: Making a puzzle of personalized e-commerce

e-Commerce recommender systems select potentially interesting products for users by looking at their purchase histories and preferences. In order to compare the available products against those included in the user’s profile, semantics-based recommendation strategies consider metadata annotations that describe their main attributes. Besides, to ensure successful suggestions of products, these strategies adapt the recommendations as the user’s preferences evolve over time. Traditional approaches face two limitations related to the aforementioned features. First, product providers are not typically willing to take on the tedious task of annotating accurately a huge diversity of commercial items, thus leading to a substantial impoverishment of the personalization quality. Second, the adaptation process of the recommendations misses the time elapsed since the user has bought an item, which is an essential parameter that affects differently to each purchased product. This results in some pointless recommendations, e.g. including regularly items that the users are only willing to buy sporadically. In order to fight both limitations, we propose a personalized e-commerce system with two main features. On the one hand, we incentivize the users to provide high-quality metadata for commercial products; on the other, we explore a strategy that offers time-aware recommendations by combining semantic reasoning about these annotations with item-specific time functions. The synergetic effects derived from this combination lead to suggestions adapted to the particular needs of the users at any time. This approach has been experimentally validated with a set of users who accessed our personalized e-commerce system through a range of fixed and handheld consumer devices.     

Understanding the effects of contextual constraints on performative behaviour in interactive media installations

In this article, we describe three interactive media installations, each evaluated in a distinctive environment. By following a research in and through design approach and studying the installations in public settings, we were able to identify an effect of contextual constraints—such as location, prominence of spectacle, length of interaction and spatial distribution of focal points—on the types of interactions encouraged through the installations. More specifically, we were able to formulate distinct content strategies for individual and group interactions while observing specific design parameters conducive to performative behaviour. We associate such parameters to three different categories of interaction with public media installations: performative interaction, ubiquitous interaction and a third hybrid scenario falling between those two, immersive interactions. We then present a framework for assessment of public interactive installations and key aspects to be considered when designing proactive contextual interventions in the public realm. Finally, we discuss how such aspects point to further investigation on formal principles underlying interactive experiences designed to facilitate specific levels of performance and spectacle.     

Assessing the influence of point-of-interest features on the classification of place categories

Points of interest (POIs) digitally represent real-world amenities as point locations. POI categories (e.g. restaurant, hotel, museum etc.) play a prominent role in several location-based applications such as social media, navigation, recommender systems, geographic information retrieval tools, and travel-related services. The majority of user queries in these applications center around POI categories. For instance, people often search for the closest pub or the best value-for-money hotel in an area. To provide valid answers to such queries, accurate and consistent information on POI categories is an essential requirement. Nevertheless, category-based annotations of POIs are often missing. The task of annotating unlabeled POIs in terms of their categories — known as POI classification — is commonly achieved by means of machine learning (ML) models, often referred to as classifiers. Central to this task is the extraction of known features from pre-labeled POIs in order to train the classifiers and, then, use the trained models to categorize unlabeled POIs. However, the set of features used in this process can heavily influence the classification results. Research on defining the influence of different features on the categorization of POIs is currently lacking. This paper contributes a study of feature importance for the classification of unlabeled POIs into categories. We define five feature sets that address operation based, review-based, topic-based, neighborhood-based, and visual attributes of POIs. Contrary to existing studies that predominantly use multi-class classification approaches, and in order to assess and rank the influence of POI features on the categorization task, we propose both a multi-class and a binary classification approach. These, respectively, predict the place category among a specified set of POI categories, or indicate whether a POI belongs to a certain category. Using POI data from Amsterdam and Athens to implement and evaluate our study approach, we show that operation based features, such as opening or visiting hours throughout the day, are the most important place category predictors. Moreover, we demonstrate that the use of feature combinations, as opposed to the use of individual features, improves the classification performance by an average of 15%, in terms of F1-score.     

基于UPnP技术设计的多媒体云服务最佳播放设备智能选取系统

当前,许多媒体服务供应商利用云技术向使用者提供流媒体云服务。云服务虽然提升了流媒体业务按需访问的便捷性,但用户在使用流媒体云服务的同时操作的智能化程度较低。用户在流媒体文件选择、媒体设备选择及服务器连接方面缺乏智能手段。此外,系统不具有媒体流播放的断点支持功能。本文利用即插即用网络通信协议UPnP,设计了一套最佳播放设备的智能选取模型。本模型通过分析比较媒体文件元数据与播放设备元数据,自动选取最佳的播放设备。同时,本模型通过断点信息的保存来实现媒体文件二次播放的连续性。本模型为流媒体云服务的断点播放和播放设备智能优化选取,提供了一种有效的技术模型。     

From MPEG-7 user interaction tools to hanging basket models: bridging the gap

The content–user gap is the difference between the limited range of content-relevant preferences that may be expressed using the MPEG-7 user interaction tools and the much wider range of metadata that may be represented using the MPEG-7 content tools. One approach for closing this gap is to make the user and content metadata isomorphic by using the existing MPEG-7 content tools to represent user (as well as content) metadata (Agius and Angelides 2006, 2007). Subsequently, user preferences may be specified for all content, without omission. Since there is a wealth of user preference and history metadata within the MPEG-7 user interaction tools that can usefully complement these specific content preferences, in this paper we develop a method by which all user and content metadata may be bridged.

DanVideo: an MPEG-7 authoring and retrieval system for dance videos

A well-annotated dance media is an essential part of a nation’s identity, transcending cultural and language barriers. Many dance video archives suffer from problems concerning authoring and access, because of the complex spatio-temporal relationships that exist between the dancers in terms of movements of their body parts and the emotions expressed by them in a dance. This paper presents a system named DanVideo for semi-automatic authoring and access to dance archives. DanVideo provides methods of annotation and authoring and retrieval tools for choreographers, dancers, and students. We demonstrate how dance media can be semantically annotated and how this information can be used for the retrieval of the dance video semantics. In particular, DanVideo offers an MPEG-7 based semi-automatic authoring tool that takes dance video annotations generated by dance experts and produces MPEG-7 metadata. DanVideo also has a search engine that takes users’ queries and retrieves dance semantics from metadata arranged using tree-embedding technique and based on spatial, temporal and spatio-temporal features of dancers. The search engine also leverages a domain-specific ontology to process knowledge-based queries. We have assessed the dance-video queries and semantic annotations in terms of precision, recall, and fidelity.     

Opinion-Based Entity Ranking using learning to rank

As social media and e-commerce on the Internet continue to grow, opinions have become one of the most important sources of information for users to base their future decisions on. Unfortunately, the large quantities of opinions make it difficult for an individual to comprehend and evaluate them all in a reasonable amount of time. The users have to read a large number of opinions of different entities before making any decision. Recently a new retrieval task in information retrieval known as Opinion-Based Entity Ranking (OpER) has emerged. OpER directly ranks relevant entities based on how well opinions on them are matched with a user's preferences that are given in the form of queries. With such a capability, users do not need to read a large number of opinions available for the entities. Previous research on OpER does not take into account the importance and subjectivity of query keywords in individual opinions of an entity. Entity relevance scores are computed primarily on the basis of occurrences of query keywords match, by assuming all opinions of an entity as a single field of text. Intuitively, entities that have positive judgments and strong relevance with query keywords should be ranked higher than those entities that have poor relevance and negative judgments. This paper outlines several ranking features and develops an intuitive framework for OpER in which entities are ranked according to how well individual opinions of entities are matched with the user's query keywords. As a useful ranking model may be constructed from many ranking features, we apply learning to rank approach based on genetic programming (GP) to combine features in order to develop an effective retrieval model for OpER task. The proposed approach is evaluated on two collections and is found to be significantly more effective than the standard OpER approach.     

Chams: Churn-aware overlay construction for media streaming

Overlay networks support a wide range of peer-to-peer media streaming applications on the Internet. The user experience of such applications is affected by the churn resilience of the system. When peers disconnect from the system, streamed data may be delayed or lost due to missing links in the overlay topology. In this paper, we explore a proactive strategy to create churn-aware overlay networks that reduce the potential of disruptions caused by churn events. We describe Chams, a middleware for constructing overlay networks that mitigates the impact of churn. Chams uses a ??hybrid?? approach??it implicitly defines an overlay topology using a gossip-style mechanism, while taking the reliability of peers into account. Unlike systems for overlay construction, Chams supports a variety of topologies used in media streaming systems, such as trees, multi-trees and forests. We evaluate Chams with different topologies and show that it reduces the impact of churn, while imposing only low computational and message overheads.     

Analysis of a CDN–P2P hybrid architecture for cost-effective streaming media distribution

To distribute video and audio data in real-time streaming mode, two different technologies – Content Distribution Network (CDN) and Peer-to-Peer (P2P) – have been proposed. However, both technologies have their own limitations: CDN servers are expensive to deploy and maintain, and consequently incur a cost for media providers and/or clients for server capacity reservation. On the other hand, a P2P-based architecture requires sufficient number of seed supplying peers to jumpstart the distribution process. Compared with a CDN server, a peer usually offers much lower out-bound streaming rate and hence multiple peers must jointly stream a media data to a requesting peer. Furthermore, it is not clear how to determine how much a peer should contribute back to the system after receiving the media data, in order to sustain the overall media distribution capacity.In this paper, we propose and analyze a novel hybrid architecture that integrates both CDN- and P2P-based streaming media distribution. The architecture is highly cost-effective: it significantly lowers the cost of CDN capacity reservation, without compromising the media quality delivered. In particular, we propose and compare different limited contribution policies for peers that request a media data, so that the streaming capacity of each peer can be exploited on a fair and limited basis. We present: (1) in-depth analysis of the proposed architecture under different contribution policies, and (2) extensive simulation results which validate the analysis. Our analytical and simulation results form a rigorous basis for the planning and dimensioning of the hybrid architecture.     

Efficient Relational Techniques for Processing Graph Queries

Graphs are widely used for modeling complicated data such as social networks, chemical compounds, protein interactions and semantic web. To effectively understand and utilize any collection of graphs, a graph database that efficiently supports elementary querying mechanisms is crucially required. For example, Subgraph and Supergraph queries are important types of graph queries which have many applications in practice. A primary challenge in computing the answers of graph queries is that pair-wise comparisons of graphs are usually hard problems. Relational database management systems (RDBMSs) have repeatedly been shown to be able to efficiently host different types of data such as complex objects and XML data. RDBMSs derive much of their performance from sophisticated optimizer components which make use of physical properties that are specific to the relational model such as sortedness, proper join ordering and powerful indexing mechanisms. In this article, we study the problem of indexing and querying graph databases using the relational infrastructure. We present a purely relational framework for processing graph queries. This framework relies on building a layer of graph features knowledge which capture metadata and summary features of the underlying graph database. We describe different querying mechanisms which make use of the layer of graph features knowledge to achieve scalable performance for processing graph queries. Finally, we conduct an extensive set of experiments on real and synthetic datasets to demonstrate the efficiency and the scalability of our techniques.