Webpage sensing information block judgment method based on visual feature extraction

The invention discloses a webpage sensing information block judgment method based on visual feature extraction, which relates to the technical field of webpage information judgment, and comprises the following steps: firstly, judging whether a target webpage conforms to a domain condition or not, an...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: ZHANG YONGXIA, LIANG SHUJUN, LIN BAODE, TIAN ERLIN
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a webpage sensing information block judgment method based on visual feature extraction, which relates to the technical field of webpage information judgment, and comprises the following steps: firstly, judging whether a target webpage conforms to a domain condition or not, and if so, entering the next step; performing visual feature denoising on the DOM tree structure of the target webpage meeting the domain conditions to obtain a text part; irrelevant information is removed from the denoised text part; carrying out region division on the residual text part to obtain a plurality of information blocks; and carrying out sensing information judgment on the plurality of information blocks. According to the method, whether the field to which the webpage belongs is the technical field applied by the sensor or not is judged firstly, then the webpage in accordance with the field is preprocessed to obtain a plurality of information blocks, and whether the webpage information blocks are the sens