Web Information retrieval Web IR Handout 1 Web
Web Information retrieval (Web IR) Handout #1: Web characteristics Ali Mohammad Zareh Bidoki ECE Department, Yazd University alizareh@yaduni. ac. ir Autumn 2011 1
Outline • • • Web challenges SE & Web IR challenges Web Structure (Graph) Web characteristics Zip law Autumn 2011 2
Web Challenges • Huge size of information – 11. 5 billions pages (2005) – 64 billions pages (05 June, 2008) • Proliferation and dynamic nature – New pages are created at the rate of 8% per week – Only 20% of the current pages will be accessible after one year – New links are created at rate 25% per week • Heterogeneous contents – HTML/Text/Audio/… • Users of web are growing exponentially Autumn 2011 3
What is the success reason of the Web? • A distributed system • A simple protocol • Production and generation is very simple Autumn 2011 4
Information Retrieval Definition • IR deals with the representation, storage, organization of, and access to information items (relevant to user query) • Information retrieval (IR) is the science of searching for documents, for information within documents, and for metadata about documents • An information retrieval process begins when a user enters a query into the system. Queries are formal statements of information needs, for example search strings in web search engines. • In information retrieval a query does not uniquely identify a single object in the collection. Instead, several objects may match the query, perhaps with different degrees of relevancy. Autumn 2011 5
Web Retrieval Search Engine User Space Matching Retrieval Browsing Information Space Index terms Full text + Structure (e. g. hypertext) Search engine is an IR system! Autumn 2011 6
IR vs Data Retrieval • A data retrieval aims at retrieving all objects which satisfy clearly defined conditions in regular expression • DR does not solve the problem of retrieving information about subject or object Autumn 2011 7
Comparing IR to databases retrieval) Databases Data Structured IR Unstructured Clear semantics No fields (other than text) Defined (relational Free text (“natural language”), Boolean Fields (SSN, age) Queries algebra, SQL) Complete Query specification Exact (results are Matching always “correct”) Error response Sensitive Autumn 2011 (vs data Incomplete Imprecise (need to measure effectiveness) Insensitive 8
Main points in IR • What is the definition of relevancy? • Evaluation! – Subjective (opposite to hardware, network) Autumn 2011 9
Web IR (SE) Challenges (1) • The definition of Relevancy • The connectivity with content in Web – A huge graph • Different type of Queries – Narrow • Needle in a haystack – Wide • Overlapping with many areas • User have Poor patience: they commonly browse through the first ten results (i. e. one screen) hoping to find there the “right” document for their query Autumn 2011 10
Web IR (SE) Challenges (2) • Spamming phenomenon – it is crucial for business sites to be ranked highly by the major search engines. – There are quite a few companies who sell this kind of expertise (also known as “search engine optimization”) and actively research ranking algorithms and heuristics of search engines, and know how many keywords to place (and where) in a Web page so as to improve the page’s ranking – SEO Books • Content & Connectivity Spamming • Anti Spamming solutions Autumn 2011 11
Web IR (SE) Challenges (3) • Rich-get-richer problem – It takes a long time for a young high quality web pages to receive an appropriate quality – Unfairness – Bad directions in growing web contents Autumn 2011 12
Web IR (SE) Challenges (4) • Crawling challenges – Huge size of information with dynamic nature – Freshness & converge • Google covers only 70% of the Web – An suitable scheduling policy – Hidden web (600 times bigger) • Using meta search engines to increase coverage – Merging and ranking problem Autumn 2011 13
Web IR (SE) Challenges (5) • User evaluation is subjective and changes in time – Relevancy between a query and document depends on user and time – Two users with the same query expect different results Autumn 2011 14
Web IR (SE) Challenges (6) • Query Ambiguity – Python – Car & automobile Autumn 2011 15
Web Dynamics • For each page p and each visit, the following information is available: – The access time-stamp of the page: visitp. – The last-modified time-stamp (given by most. Web servers; about 80%-90%of the requests in practice): modifiedp. – The text of the page, which can be compared to an older copy to detect changes, especially if modifiedp – is not provided. – The following information can be estimated if the re-visiting period is short: – The time at which the page first appeared: createdp. – The time at which the page was no longer reachable: deletedp • In all cases, the results are only an estimation of the actual values Autumn 2011 16
Estimating freshness and age • The probability that a copy of p is up-todate at time t, up(t) decreases with time if the page is not re-visited. • When page changes are modeled as a Poisson process, if t units of time have passed since the last visit, then: Autumn 2011 17
Characterization of Web page changes • Age: visitp-modifiedp. • Lifespan: deletedp-createdp. • Number of changes during the lifespan: changesp. • Average change interval: lifespanp/changesp. Autumn 2011 18
Freshness && Age Autumn 2011 19
Autumn 2011 20
Web a Scale Free Network • A scale-free network is characterized by a few highly-linked nodes that act as “hubs” connecting several nodes to the network. • It follows Power Law Autumn 2011 21
Random Vs Scale-Free Autumn 2011 22
Distribution of Web Graph: Power. Law Autumn 2011 23
Power-Law and Zipf Law Autumn 2011 24
Zipf Law for Content Autumn 2011 25
Macroscopic Structure of Web Autumn 2011 26
User Sessions • User sessions on the Web are usually characterized through models of random surfers • The most used source for data about the browsing activities of users are the access log files of Web Servers, Proxies, SEs – Caching • Modeling User behavior • Eye tracking Autumn 2011 27
Next Lecture • Information Retrieval Models – Boolean – Vector Space – Realistic Autumn 2011 28
- Slides: 28