Document recommending method, device, equipment and computer readable medium
A recommendation method and document technology, applied in computing, special data processing applications, instruments, etc., can solve problems such as user flooding, and achieve the effect of accurate document recommendation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0080] Such as figure 1 As described, the document recommendation method in this embodiment includes:
[0081] S110, acquiring the original document.
[0082] Among them, the original document includes a title and a text, and the topic information can be extracted from the title, that is, the field to which the original document belongs, such as science and technology, finance, military, etc., and the focus information can be extracted from the text, such as keywords, named entities, etc.
[0083] In one embodiment, the original document may include the document being clicked by the user, and may also include the document involved in the user's click log, wherein the user click log includes data of historically clicked documents within a set time. Wherein, the set time includes any time range before the current time, which can be selected according to the characteristics of different users or the requirements of actual application scenarios. For example, the document click r...
Embodiment 2
[0103] This embodiment provides a document recommendation method, which further includes before step S120 described in Embodiment 1:
[0104] Step S210, acquiring training corpus. Wherein, the training corpus includes a plurality of sample documents and classification data marked on the sample documents.
[0105] Step S220, based on the training corpus, perform supervised learning model training to obtain a pan-label classification model.
[0106] In an implementable manner, step S210, obtaining training corpus, may be performed based on a pan-label classification system, such as Figure 5 As shown, the acquisition of training corpus described in step S210 includes:
[0107]Step S211, establishing a pan-label classification system. Among them, the pan-label classification system can be established by receiving classification rules from outside. The external classification rules can be determined by the editor in combination with the field experience of each vertical catego...
Embodiment 3
[0121] An embodiment of the present invention provides a device for recommending documents, corresponding to the method for recommending documents in Embodiment 1. Such as Figure 8 mentioned, including:
[0122] The original document obtaining module 110 is used to obtain the original document, wherein the original document includes a title and a text, which are respectively used to extract topic information and focus information;
[0123] The pan-tag determining module 120 is configured to determine the pan-tag of the original document by using a pan-tag classification model, wherein the classification granularity of the pan-tag is between the classification granularity based on the topic class information and the classification granularity based on the focus information between the classification granularities of ; and
[0124] The recommended document determining module 130 is configured to determine a recommended document, wherein the recommended document has a matching...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com