Time Objects for Chronological Information Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional web search systems face challenges in providing relevant, time-pertinent information in a fast and computationally inexpensive manner, requiring users to manually sift through ranked lists of web pages, which can be time-consuming and complex.
Innovation Solution
The implementation of time objects, which are composites of content from different media types related to a specific date or date range, created by a web crawler and displayed in a time-based manner, allowing users to query and retrieve information intuitively through a time search engine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional web search systems rank web pages by relevance, then users can find relevant information, but users must manually sift through ranked lists which is time-consuming and complex
Solution Approach 1:
The patent segments information retrieval by time periods, dividing the search results into distinct temporal categories (e.g., recent, past year, past decade). This segmentation allows users to quickly navigate to specific time ranges without manually filtering through all results, directly reducing the time and effort required to find time-pertinent information.
Solution Approach 2:
The patent introduces time as an additional dimension for organizing and displaying search results, beyond traditional relevance ranking. By presenting results in a time-based dimension (chronological order, time periods, temporal trends), users can intuitively understand when information was created or updated, making it easier to identify time-pertinent information without manual effort.
2Quantity of substance
If web crawlers index billions of web pages, then comprehensive information is available, but storage resources are heavily consumed
Solution Approach 1:
The patent extracts and stores only the essential time-related metadata (creation date, update date, publication date) from web pages during crawling, rather than storing complete page content. This extraction approach maintains comprehensive temporal information for billions of pages while significantly reducing storage requirements, as only key temporal attributes are retained for time-based querying and filtering.
3Reliability
If information is organized by traditional keywords, then search relevance is improved, but time-pertinence of information is not adequately addressed
Solution Approach 1:
The patent merges traditional keyword-based relevance ranking with time-based organization into a unified search system. Both keyword matching and temporal metadata are integrated to determine search results, allowing the system to simultaneously evaluate content relevance and time-pertinence. This combination ensures that users receive results that are both relevant to their query and appropriate for the desired time period.
Data Source
AI summary
Information retrieval using time is described. In an embodiment a web-crawler creates time objects which are composites of content of different media types obtained from potentially different sources and, for example, are about the same date or date range. For example, a time object may comprise an image and a piece of text about an event in 1977. In this case the time object may have the date 1977 stored with it as well as URLs of originating web pages. In an embodiment a user is able to query a search engine to retrieve time objects which are displayed in a manner related to the dates of the time objects. For example, the time objects may be displayed in chronological order against a time line. In examples time objects may be dynamically created as a web page is requested and displayed with that web page.