Intelligent Search from Yahoo's experiment could predict future events

yahoosearch.jpg

Semantic and contextualized search is an important area which will impact our ability to quickly do “intelligent” searches around the web.  As the information on web has grown – one came to realize that a vast amount of it was lying there with no context in the direct sense.  When such information is searched just for key words, then one could get results that aren’t always useful or effective.  But if one could “attach” context to the information, which is relevant to the person searching, suddenly the results returned have a far greater value.

For example, if I am searching for “Energy” in a scientific context and want to know its formulas or science behind it, Google’s search results may not be the most efficient.  So in came Wolfram Search.  It contextualizes search words – by asking you if you want to search it as a ‘Word”, formula or a physical quantity.  The results are different and useful for the context they are used by the person searching.  Compared to Google search this goes several notches higher in terms of search technology.

Some of the areas of Wolfram search include:

Mathematics » Elementary Math   Numbers   Plotting   Algebra   Matrices   Calculus   Geometry   Trigonometry   Discrete Math   Number Theory   Applied Math   Logic   Functions
Statistics & Data Analysis » Descriptive Statistics   Regression   Statistical Distributions   Probability  
Physics » Mechanics   Electricity & Magnetism   Optics   Relativity   Nuclear Physics   Quantum Physics   Particle Physics   Statistical Physics   Astrophysics   Physical Constants  
Chemistry » Elements   Compounds   Ions   Quantities   Solutions   Reactions   Chemical Thermodynamics
Materials » Alloys   Minerals   Plastics   Woods   Bulk Materials   …
Engineering » Acoustics   Aeronautics   Electric Circuits   Fluid Mechanics   Steam Tables   Structures
Astronomy » Star Charts   Astronomical Events   Planets   Moons   Minor Planets   Comets   Space Weather   Stars   Pulsars   Galaxies   Star Clusters   Nebulae   Astrophysics   .
Food & Nutrition » Foods   Dietary References
Words & Linguistics » Dictionary Lookup   Anagrams   Word Puzzles   Morse Code   Soundex   Languages   Number Names   …
Culture & Media » Books   Periodicals   Movies   Fictional Characters   Television Networks   Songs   Awards   …
People & History » People   Genealogy   Names   Occupations   Political Leaders   Historical Events   Historical Periods   Historical Countries   Historical Numerals   Historical Money
Education »Universities   Standardized Tests   …
Organizations » Universities   Companies   Hospitals   Foundations   International Organizations   …
Sports & Games » Football   Baseball   Olympics   Lotteries   Card Games   …
Music »  Musical Notes   Intervals   Chords   Scales   Songs   …
Colors » Color Names   Color Addition   Color Systems   Temperatures   Wavelengths

There is another experiment in contextual search by ZCubes.com – where they have layered the Google search (and other search engines) to serve contextually relevant search results.  Depending on who you are and what you are really looking for, you can get different search results.

Now, Yahoo has entered this world of intelligent searches via an experimental project called Time Explorer.  This project is to be in conjunction with the broader Livingknowledge initiative.  The site officially explains it like this:

The application will be eventually be a showcase for the functionality of the LivingKnowledge project. This current version is designed to demonstrate the current state of the project using the NYT collection as part of the HCIR challenge

What this experiment primarily does is to contextualize news items by time.  Time, they believe, is an important element and if a news story can be followed across time then a better context emerges.  It can also be used to predict the future.  So, one goes into the past and then uses it to make future predictions.  That, is what this new type of search engine tries to do.

In current news search engines, time is primarily used to boost the relevance of the most recent stories. While useful when users are interested in the latest news, it may hinder the search experience of those interested in a broader understanding of a particular news story. These users may benefit from a transversal organization of the topic across time so as to better view how the story has evolved and which people and places have shaped the evolution. Furthermore, these users may equally benefit from predictions on how the story might evolve into the future. When searching about a regional conflict, for example, a user should be able to identify what factors lead to the conflict, which people where most influential and when, and how the conflict is likely to evolve in the future.

Search Results are shown on a timeline scale which stretches several years back and forth. To use it, just move your mouse over the future timeline scale, and you can view the predictions for what was supposed to happen in that year given the past occurrences from as far back as 20 years ago.

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to Drishtikone - Online Magazine on Geopolitics and Culture from Indian Perspective.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.