Web-scale editorial crawling must balance coverage and freshness within tight politeness and request budgets. Nonetheless, in production systems, manual seed management remains common despite being inefficient—oversampling redundant seeds while missing high-yield ones. URLBank replaces manual curation with a label-free controller that infers optimal seed selection directly from temporal crawl telemetry. It identifies candidate entry points, estimates stability—the persistence of links across crawler revolutions—and productivity—the rate of first-seen publications—and ranks them through greedy marginal gain on a shared-credit coverage objective. In a shadow A/B evaluation spanning 5,238 sites, URLBank consistently achieves higher coverage, greater efficiency, and earlier discovery under identical conditions. Gains remain stable across Top-K budgets, approaching near-complete coverage with far fewer seeds. Deployed alongside Meltwater's production crawler Pulitzer, URLBank operates with versioned policies, ranked prefixes for crawl budgets, and integrated health diagnostics, making allocation transparent, auditable, and reversible. Together, these results demonstrate that temporal signals, through an interpretable greedy objective, yield large, measurable improvements in industrial-scale coverage, resource efficiency, and freshness.

Data-driven websites are mostly accessed through search interfaces. Such sites follow a common publishing pattern that, surprisingly, has not been fully exploited for unsupervised data extraction yet: the result of a search is presented as a paginated list of result records. Each result record contains the main attributes about one single object, and links to a page dedicated to the details of that object.

Most modern web scrapers use an embedded browser to render web pages and to simulate user actions. Such scrapers (or wrappers) are therefore expensive to execute, in terms of time and network traffic. In contrast, it is magnitudes more resource-efficient to use a "browserless" wrapper which directly accesses a web server through HTTP requests, and takes the desired data directly from the raw replies. However, creating and maintaining browserless wrappers of high precision requires specialists, and is prohibitively labor-intensive at scale. In this paper, we demonstrate the principal feasibility of automatically translating browser-based wrappers into "browserless" wrappers. We present the first algorithm and system performing such an automated translation on suitably restricted types of web sites. This system works in the vast majority of test cases and produces very fast and extremely resource-efficient wrappers. We discuss research challenges for extending our approach to a general method applicable to a yet larger number of cases.

Aspect extraction identifies relevant features from a textual description of an entity, e.g., a phone, and is typically targeted to product descriptions, reviews, and other short texts as an enabling task for, e.g., opinion mining and information retrieval. Current aspect extraction methods mostly focus on aspect terms and often neglect interesting modifiers of the term or embed them in the aspect term without proper distinction. Moreover, flat syntactic structures are often assumed, resulting in inaccurate extractions of complex aspects. This paper studies the problem of structured aspect extraction, a variant of traditional aspect extraction aiming at a fine-grained extraction of complex (i.e., hierarchical) aspects. We propose an unsupervised and scalable method for structured aspect extraction consisting of statistical noun phrase clustering, cPMI-based noun phrase segmentation, and hierarchical pattern induction. Our evaluation shows a substantial improvement over existing methods in terms of both quality and computational efficiency.

Data wrangling is the process by which the data required by an application is identified, extracted, cleaned and integrated, to yield adata set that is suitable for exploration and analysis. Although there are widely used Extract, Transform and Load (ETL) techniques and platforms, they often require manual work from technical and domain experts at different stages of the process. When confronted with the 4 V’s of big data (volume, velocity, variety and veracity),manual intervention may make ETL prohibitively expensive. This paper argues that providing cost-effective, highly-automated approaches to data wrangling involves significant research challenges,requiring fundamental changes to established areas such as data extraction,integration and cleaning, and to the ways in which these areas are brought together. Specifically, the paper discusses the importance of comprehensive support for context awareness within data wrangling, and the need for adaptive, pay-as-you-go solutions that automatically tune the wrangling process to the requirements and resources of the specific application.

Automated web scraping is a popular means for acquiring data from the web. Scrapers (or wrappers) are derived from either manually or automatically annotated examples, often resulting in under/over segmented data, together with missing or spurious content. Automatic repair and maintenance of the extracted data is thus a necessary complement to automatic wrapper generation. Moreover, the extracted data is often the result of a long-term data acquisition effort and thus jointly repairing wrappers together with the generated data reduces future needs for data cleaning. We study the problem of computing joint repairs for XPath-based wrappers and their extracted data. We show that the problem is NP-complete in general but becomes tractable under a few natural assumptions. Even tractable solutions to the problem are still impractical on very large datasets, but we propose an optimal approximation that proves effective across a wide variety of domains and sources. Our approach relies on encoded domain knowledge, but require no per-source supervision. An evaluation spanning more than 100k web pages from 100 different sites of a wide variety of application domains, shows that joint repairs are able to increase the quality of wrappers between 15% and 60% independently of the wrapper generation system, eliminating all errors in more than 50% of the cases.

Wrapper induction is the problem of automatically inferring a query from annotated web pages of the same template. This query should not only select the annotated content accurately but also other content following the same template. Beyond accurately matching the template, we consider two additional requirements: (1) wrappers should be robust against a large class of changes to the web pages, and (2) the induction process should be noise resistant, i.e., tolerate slightly erroneous (e.g., machine generated) samples. Key to our approach is a query language that is powerful enough to permit accurate selection, but limited enough to force noisy samples to be generalized into wrappers that select the likely intended items. We introduce such a language as subset of XPATH and show that even for such a restricted language, inducing optimal queries according to a suitable scoring is infeasible. Nevertheless, our wrapper induction framework infers highly robust and noise resistant queries. We evaluate the queries on snapshots from web pages that change over time as provided by the Internet Archive, and show that the induced queries are as robust as the human-made queries. The queries often survive hundreds sometimes thousands of days, with many changes to the relative position of the selected nodes (including changes on template level). This is due to the few and discriminative anchor (intermediately selected) nodes of the generated queries. The queries are highly resistant against positive noise (up to 50%) and negative noise (up to 20%).

The web is the largest bulletin board of the world. Events of all types, from flight arrivals to business meetings, are announced on this board. Tracking and reacting to such event announcements, however, is a tedious manual task, only slightly alleviated by email or similar notifications. Announcements are published with human readers in mind, and updates or delayed announcements are frequent. These characteristics have hampered attempts at automatic tracking. P ea CE provides the first integrated framework for event processing on top of web event ads, consisting of event extraction, complex event processing, and action execution in response to these events. Given a schema of the events to be tracked, the framework populates this schema by extracting events from announcement sources. This extraction is performed by little programs called wrappers that produce the events including updates and retractions. P ea CE then queries these events to detect complex events, often combining announcements from multiple sources. To deal with updates and delayed announcements, P ea CE’s schemas are bitemporal, to distinguish between occurrence and detection time. This allows complex event specifications to track updates and to react upon differences in occurrence and detection time. In case of new, changing, or deleted events, P ea CE allows one to execute actions, such as tweeting or sending out email notifications. Actions are typically specified as web interactions, for example, to fill and submit a form with attributes of the triggering event. Our evaluation shows that P ea CE’s processing is dominated by the time needed for accessing the web to extract events and perform actions, allotting to 97.4%. Thus, P ea CE requires only 2.6% overhead, and therefore, the complex event processor scales well even with moderate resources. We further show that simple and reasonable restrictions on complex event specifications and the timing of constituent events suffice to guarantee that P ea CE only requires a constant buffer to process arbitrarily many event announcements.

Web scraping (or wrapping) is a popular means for acquiring data from the web. Recent advancements have made scalable wrapper-generation possible and enabled data acquisition processes involving thousands of sources. This makes wrapper analysis and maintenance both needed and challenging as no scalable tools exists that support these tasks. We demonstrate WADaR, a scalable and highly automated tool for joint wrapper and data repair. WADaR uses off-the-shelf entity recognisers to locate target entities in wrapper-generated data. Markov chains are used to determine structural repairs, that are then encoded into suitable repairs for both the data and corresponding wrappers. We show that WADaR is able to increase the quality of wrapper-generated relations between 15% and 60%, and to fully repair the corresponding wrapper without any knowledge of the original website in more than 50% of the cases.

The web is overflowing with implicitly structured data, spread over hundreds of thousands of sites, hidden deep behind search forms, or siloed in marketplaces, only accessible as HTML. Automatic extraction of structured data at the scale of thousands of websites has long proven elusive, despite its central role in the "web of data". Through an extensive evaluation spanning over 10000 web sites from multiple application domains, we show that automatic, yet accurate full-site extraction is no longer a distant dream. diadem is the first automatic full-site extraction system that is able to extract structured data from different domains at very high accuracy. It combines automated exploration of websites, identification of relevant data, and induction of exhaustive wrappers. Automating these components is the first challenge. diadem overcomes this challenge by combining phenomenological and ontological knowledge. Integrating these components is the second challenge. diadem overcomes this challenge through a self-adaptive network of relational transducers that produces effective wrappers for a wide variety of websites. Our extensive and publicly available evaluation shows that, for more than 90% of sites from three domains, diadem obtains an effective wrapper that extracts all relevant data with 97% average precision. diadem also tolerates noisy entity recognisers, and its components individually outperform comparable approaches.

Forms are our gates to the web. They enable us to access the deep content of web sites. Automatic form understanding unlocks this content for applications ranging from crawlers to meta-search engines and is essential for improving usability and accessibility of the web. Form understanding has received surprisingly little attention other than as component in specific applications such as crawlers. No comprehensive approach to form understanding exists and previous works disagree even in the definition of the problem. In this paper, we present OPAL, the first comprehensive approach to form understanding. We identify form labeling and form interpretation as the two main tasks involved in form understanding. On both problems OPAL pushes the state of the art: For form labeling, it combines signals from the text, structure, and visual rendering of a web page, yielding robust characterisations of common design patterns. In extensive experiments on the ICQ and TEL-8 benchmarks and a set of 200 modern web forms OPAL outperforms previous approaches by a significant margin. For form interpretation, we introduce a template language to describe frequent form patterns. These two parts of OPAL combined yield form understanding with near perfect accuracy (> 98%).

Although deep web analysis has been studied extensively, there is no succinct formalism to describe user interactions with AJAX-enabled web applications.

Web crawlers are increasingly used for focused tasks such as the extraction of data from Wikipedia or the analysis of social networks like last.fm. In these cases, pages are far more uniformly structured than in the general Web and thus crawlers can use the structure of Web pages for more precise data extraction and more expressive analysis.

Applications and services that access Web data are becoming increasingly more useful and wide-spread. Current main-stream Web query languages such as XQuery, XSLT, or SPARQL, however, focus only on one of the different data formats available on the Web. In contrast, Xcerpt is a emphversatile semi-structured query language, i.e., a query language able to access all kinds of Web data such as XML and RDF in the same language reusing common concepts and language constructs. To integrate heterogeneous data and as a foundation for Semantic Web reasoning, Xcerpt also provides rules. Xcerpt has a visual companion language, visXcerpt, that is conceived as a mere rendering of the (textual) query language Xcerpt using a slightly extended CSS. Both languages are demonstrated along a realistic use case integrating XML and RDF data highlighting interesting and unique features. Novel language constructs and optimization techniques are currently under investigation in the Xcerpt project ([email protected] urlhttp://xcerpt.org/).

Data streams are an emerging technology for data dissemination in cases where the data throughput or size makes it unfeasible to rely on the conventional approach based on storing the data before processing it. SPEX evaluates XPath queries against XML data streams. SPEX is built upon formal frameworks for (1) rewriting XPath queries into equivalent XPath queries without reverse axes and (2) correct query evaluation with polynomial combined complexity using networks of pushdown transducers. Such transducers are simple, independent, and can be connected in a flexible manner, thus allowing not only easy extensions but also extensive query optimization. Querying XML streams with SPEX consists in four steps: first, the input XPath query is rewritten into an XPath query without reverse axes. Second, the forward XPath query is compiled into a logical query plan abstracting out details of the concrete XPath syntax. Then, a physical query plan is generated by extending the logical query plan with operators for determination and collection of answers. In the last step, the XML stream is processed continuously with the physical query plan, and the output stream conveying the answers to the original query is generated progressively.

A decade of experience with research proposals as well as standardized query languages for the conventional Web and the recent emergence of query languages for the Semantic Web call for a reconsideration of design principles for Web and Semantic Web query languages. This chapter first argues that a new generation of versatile Web query languages is needed for solving the challenges posed by the changing Web: We call versatile those query languages able to cope with both Web and Semantic Web data expressed in any (Web or Semantic Web) markup language. This chapter further suggests that well-known referential transparency and novel answer-closedness are essential features of versatile query languages. Indeed, they allow queries to be considered like forms and answers like form-fillings in the spirit of the query-by-example paradigm. This chapter finally suggests that the decentralized and heterogeneous nature of the Web requires incomplete data specifications (or incomplete queries) and incomplete data selections (or incomplete answers); the form-like query can be specified without precise knowledge of the queried data, and answers can be restricted to contain only an excerpt of the queried data.