Autocomplete Search System for Small Screen Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search methods on small screen devices are inefficient, requiring multiple clicks and scrolls to find information, and assume user familiarity with navigation hierarchies, limiting accessibility and speed.
Innovation Solution
A client-server system that preprocesses websites to extract and index possible answer entities, providing a dropdown list of relevant answers as the user types, allowing direct navigation to specific portions of web pages, reducing the need for extensive scrolling and improving search speed and intuitiveness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional menu link driven navigation is used, then structured information access is provided, but the number of clicks and scrolls increases significantly
Solution Approach 1:
The system pre-processes website content to extract entities, relationships, and answers before user queries are submitted. This preliminary indexing of content metadata enables direct navigation to relevant information without requiring users to traverse multiple menu levels or scroll through pages.
Solution Approach 2:
The invention introduces an intermediary layer between the user and website content - a search system that processes queries against pre-extracted entities and relationships. This intermediary translates user intent into direct navigation paths, eliminating the need for traditional hierarchical navigation through menus and links.
2Quantity of substance
If comprehensive website content is made accessible, then information completeness improves, but screen real estate requirements increase
Solution Approach 1:
The system extracts specific entities, relationships, and answers from comprehensive website content, separating only the relevant information needed to answer user queries. This extraction approach makes substantial information accessible without displaying the entire website content, thus conserving screen real estate while maintaining information completeness.
3Loss of information
If traditional search results with multiple paragraphs are displayed, then comprehensive coverage is achieved, but user scanning time increases
Solution Approach 1:
The system segments search results into discrete, ranked answer entities rather than presenting continuous paragraphs of text. Each answer entity represents a distinct piece of information that directly addresses aspects of the user's query, enabling rapid scanning and selection without reading through multiple paragraphs.
Data Source
AI summary
In one embodiment, a server-side process operating on a server includes crawling of one or more websites for possible answers to a query and adding the possible answer entities into a data structure of answer entities to queries. The server-side process also includes serving a requested Web page with an added search box that is displayed by a client coupled to the server. As a user types one or more characters at the client, the client sends the characters, portion-by-portion, e.g., character-by-character to the server-side process. In response, the server-side process searches the index for possible answers to the (so-far-typed) query, and provides to the client a list of answers (rather than queries) to be displayed by the client. Each additional character entered may modify the provided list of answers. The client selecting a listed answer navigation directly to the answer.