Matrix Factorization Techniques for Recommender Systems

We investigate the problem of processing a large amount of continuous spatial-keyword queries over streaming data, which is essential in many applications such as location-based recommendation and advertising, thanks to the proliferation of geo-equipped devices and the ensuing location-based social media applications. For example, a location-based e-coupon system may allow potentially millions of users to register their continuous spatial-keyword queries (e.g., interests in nearby sales) by specifying a set of keywords and a spatial region; the system then delivers each incoming spatial-textual object (e.g., a geo-tagged e-coupon) to all the matched queries (i.e., users) whose spatial and textual requirements are satisfied. While there are several prior approaches aiming at providing efficient query processing techniques for the problem, their approaches belong to spatial-first indexing method which cannot well exploit the keyword distribution. In addition, their textual filtering techniques are built upon simple variants of traditional inverted indexes, which do not perform well for the textual constraint imposed by the problem. In this paper, we address the above limitations and provide a highly efficient solution based on a novel adaptive index, named AP-Tree. The AP-Tree adaptively groups registered queries using keyword and spatial partitions, guided by a cost model. The AP-Tree also naturally indexes ordered keyword combinations. We present index construction algorithm that seamlessly and effectively integrates keyword and spatial partitions. Consequently, our method adapts well to the underlying spatial and keyword distributions of the data. Our extensive experiments demonstrate that AP-Tree achieves up to an order of magnitude improvement on efficiency compared with prior state-of-the-art methods.

/pdf/ap-tree-efficiently-support-continuous-spatial-keyword-2tab2f0nee.pdf

AP-Tree: Efficiently support continuous spatial-keyword queries over stream

Techniques for performing non-event pattern matching on continuous event streams using variable duration. The duration value used in non-event pattern matching can be variable. Accordingly, a first pattern match candidate can have a different associated duration from a second pattern match candidate for matches arising from events received via an event stream. In certain embodiments, the duration for a candidate pattern match may be based upon one or more attributes of an event that started the candidate pattern match or based upon an expression (e.g., an arithmetic expression) involving one or more attributes of the event.

Variable duration non-event pattern matching

Techniques for partitioning data streams are provided. In some examples, a query for processing at least a portion of a data stream may be identified. The data stream may be associated with a user. Additionally, in some examples, code identifying an attribute of the identified stream may be received. The code may be capable of configuring the query based at least in part on the attribute. Further, in some aspects, the code may be configured to partition the data stream into at least a sub-stream based at least in part on the attribute.

Custom partitioning of a data stream

Techniques for extending the capabilities of an event processing system to support the processing of spatial data. In one set of embodiments, this extensibility can be provided via a plug-in extension component referred to herein as a spatial data cartridge. The spatial data cartridge can enable the event processing system to support spatial data types (e.g., point, polygon, etc.) and various operations related to such data types (e.g., proximity determinations, overlap determinations, etc.). The spatial data cartridge can also define an indexing scheme that can be integrated with the capabilities of the event processing system to support the indexing of spatial data. Using the spatial data cartridge, the event processing system can operate on spatial data even if spatial data formats are not natively supported by the system.

Spatial data cartridge for event processing systems

In this paper we present RoSeS (Really Open Simple and Efficient Syndication), a generic framework for content-based RSS feed querying and aggregation. RoSeS is based on a data-centric approach, using a combination of standard database concepts like declarative query languages, views and multiquery optimization. Users create personalized feeds by defining and composing content-based filtering and aggregation queries on collections of RSS feeds. Publishing these queries corresponds to defining views which can then be used for building new queries / feeds. This naturally reflects the publish-subscribe nature of RSS applications. The contributions presented in this paper are a declarative RSS feed aggregation language, an extensible stream algebra for building efficient continuous multi-query execution plans for RSS aggregation views, a multiquery optimization strategy for these plans and a running prototype based on a multi-threaded asynchronous execution engine.

/pdf/roses-a-continuous-content-based-query-engine-for-rss-feeds-1dhabtpwtm.pdf

RoSeS: a continuous content-based query engine for RSS feeds

We present the WebContent platform for managing distributed repositories of XML and semantic Web data. The platform allows integrating various data processing building blocks (crawling, translation, semantic annotation, full-text search, structured XML querying, and semantic querying), presented as Web services, into a large-scale efficient platform. Calls to various services are combined inside ActiveXML [8] documents, which are XML documents including service calls. An ActiveXML optimizer is used to: (i) efficiently distribute computations among sites; (ii) perform XQuery-specific optimizations by leveraging an algebraic XQuery optimizer; and (iii) given an XML query, chose among several distributed indices the most appropriate in order to answer the query.

/pdf/webcontent-efficient-p2p-warehousing-of-web-data-4or5m1x0ey.pdf

WebContent: efficient P2P Warehousing of web data

The explosion of published information on the Web leads to the emergence of a Web syndication paradigm, which transforms the passive reader into an active information collector. Information consumers subscribe to RSS/Atom feeds and are notified whenever a piece of news (item) is published. The success of this Web syndication now offered on Web sites, blogs, and social media, however raises scalability issues. There is a vital need for efficient real-time filtering methods across feeds, to allow users to follow effectively personally interesting information. We investigate in this paper three indexing techniques for users' subscriptions based on inverted lists or on an ordered trie. We present analytical models for memory requirements and matching time and we conduct a thorough experimental evaluation to exhibit the impact of critical workload parameters on these structures.

/pdf/subscription-indexes-for-web-syndication-systems-48tr4ptd1l.pdf

Subscription indexes for web syndication systems

We are witnessing a widespread of web syndication technologies such as RSS or Atom for a timely delivery of frequently updated Web content. Almost every personal weblog, news portal, or discussion forum employs nowadays RSS/Atom feeds for enhancing pull-oriented searching and browsing of web pages with push-oriented protocols of web content. Social media applications such as Twitter or Facebook also employ RSS for notifying users about the newly available posts of their preferred friends. Unfortunately, previous works on RSS/Atom statistical characteristics do not provide a precise and updated characterization of feeds' behavior and content, characterization which can be used to successfully benchmark effectiveness and efficiency of various RSS processing/ analysis techniques. In this paper, we present the first thorough analysis of three complementary features of real-scale RSS feeds, namely, publication activity, items structure and length, as well as, vocabulary of its content which we believe are crucial for Web 2.0 applications.

/pdf/characterizing-web-syndication-behavior-and-content-2szwqc9jy5.pdf

Characterizing web syndication behavior and content

Tree Pattern Queries [7,6] are now well admitted for modeling parts of XML Queries. Actual works only focus on a small subpart of XQuery specifications and are not well adapted for evaluation in a distributed heterogeneous environment.

In this paper, we propose the TGV (Tree Graph View) model for XQuery processing. The TGV model extends the Tree Pattern representation in order to make it intuitive, has support for full untyped-XQuery queries, and for optimization and evaluation. Several types of Tree Pattern are manipulated to handle all XQuery requirements. Links between Tree Patterns are called hyperlinks in order to apply transformations on results.

The TGV has been implemented in a mediator system called XLive.

https://hal.archives-ouvertes.fr/hal-00733459/document

Nicolas Travers

Papers

RoSeS: a continuous content-based query engine for RSS feeds

WebContent: efficient P2P Warehousing of web data

Subscription indexes for web syndication systems

Characterizing web syndication behavior and content

TGV: a tree graph view for modeling untyped XQuery