IDS 200 Lecture Notes - Lecture 18: User Friendly, Dark Web, Big Data
12 Feb 2020
School
Department
Course
Professor
Document Summary
Intuitively, less common and more precise words will more closely match a topic: Truly relevant matches nearer to the top. Unemployment statistics will mostly return recent federal data and interpretations of data & trends. Unemployment statistics 1982 will return early 1980"s data and articles about that recession. Unemployment statistics rockford il 1982 will return several articles about that local economy. In its early days, the internet could operate like an oversized message board. Typical users want to quickly get good search results without thinking too hard. Irrelevant results waste users" time, lead to search engine abandonment. Correct relevance assessment requires additional data beyond simple term matches. Pre-www, internet search was a lot like searching for a particular file on a computer. Based on matching names, types, or metadata (e. g. , when the file was created or updated) As the world wide web dominated usage, search needed to match terms to text and content.
