Accelerating Structured Web Crawling without Losing Data

Abstract

Size of retrieved data versus crawling time formulate a well-known dilemma in the structured Web crawling community. The real challenge within this dilemma is to optimize the settings of a given wrapper to obtain maximum available data in shortest possible time. In this paper, we try to tune these settings, by introducing a threaded algorithm that guarantees accessing all available detail pages within crawling scope; and using this algorithm, we try to reduce the time consumed by the crawler, via simple adjustments of sleeping time after each detail page visit.