Contextual Book Recommendation Text for E-Book Reader Engagement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing book recommendation methods in e-book reading applications are monotonous and ineffective in attracting readers to continue reading, leading to unsatisfactory engagement.
Innovation Solution
A method for generating book recommendation texts by determining a target book based on related features of the currently read book, such as author, content, and reader comments, and inserting the recommendation text at relevant locations in the e-book, enhancing relevance and pertinence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If book recommendation is implemented by displaying covers of several other books at the last page or at the end of the current book, then the recommendation can be shown to readers, but the recommendation method becomes monotonous and could not effectively attract the readers to continue to read
Solution Approach 1:
The patent applies local quality by inserting recommendation texts at specific local positions within the book content (such as after chapters or sections) rather than only at the end. This creates differentiated recommendation placements that adapt to different reading contexts, making the recommendation less monotonous and more engaging for readers at various stages of reading.
Solution Approach 2:
The patent transitions from a single-dimensional recommendation display (only at the last page) to a multi-dimensional approach by inserting recommendations at multiple positions throughout the book structure. This dimensional expansion allows recommendations to appear at chapter endings, section breaks, and other strategic locations, thereby increasing reader engagement and effectiveness.
2Reliability
If book recommendation text is inserted at relevant locations in the e-book, then the relevance and pertinence of recommendation is enhanced, but the device complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-defining multiple insertion positions and templates for recommendation texts before the actual recommendation process. The system prepares a set of predetermined locations (such as after specific chapters or sections) and corresponding text templates in advance, so that when recommendations need to be inserted, they can be directly placed at these pre-prepared positions without complex real-time calculations, thereby reducing system complexity while maintaining high relevance.
Solution Approach 2:
The patent uses parameter changes by dynamically selecting different insertion positions and text templates based on book characteristics and reader profiles. Instead of using a fixed complex insertion mechanism, the system adjusts parameters such as insertion location and text content according to predefined criteria, simplifying the overall system architecture while achieving high recommendation relevance through parameter optimization rather than structural complexity.
Data Source
AI summary
Embodiments of the present disclosure provide a method for book pushing, a method for generating a book recommendation text, an apparatus, and an electronic device. The method for book pushing comprises: obtaining a first book; determining, according to related features of the first book, a target book in at least one second book, the second book and the first book being written by the same author; determining, according to contents of the target book and/or the first book, a recommendation text; generating, according to the recommendation text and related information of the target book, a book recommendation text of the target book; and pushing the book recommendation text to the reader. The method establishes, according to the related features of the first book currently read by the reader, correlations with the target book, and generates the book recommendation text containing the recommendation text.