Browse Prior Art Database

A cognitive search system to generate combined results by multiple dimension search Disclosure Number: IPCOM000247104D
Publication Date: 2016-Aug-05
Document File: 10 page(s) / 94K

Publishing Venue

The Prior Art Database


There is no existing search system, in one search request, can search different objects from different web pages and calculate each search result and generate combined results according to multiple dimension conditions. For example: When a consumer goes to a place to consume a consumption, generally, there might be also other expense involved nearby the same location, which can be called a group of consumption nearby a location. Currently, it is difficult for a consumer to make a decision to go to a better location by searching total cost of a group of consumption on-site: 1. Information required is in different web pages in different websites, user needs to open multiple application/web site to get them one by one and then filtering, calculating, aggregating by manual; 2. The relationship among multiple consumption in keywords are not defined and considered, such as I want to search the second consumption and the third consumption which with a condition "location nearby the first consumption", and with a condition of "lowest group price" 3. There is no way to search a lowest group price and associated with other cost (such as the cost of transportation) 4. There is no way to search a lowest group price which include personal favorite consumption This idea provides a cognitive search system which can search multiple objects from multiple 3rd party systems based on the understanding of multiple dimension relationships between each other, and have below points: 1. There are more than one objects in keywords, which are associated by multiple dimension conditions 2. This search system can generate personal favorite extension search according to input objects 3. This search system executes multiple group search, and in each group search, different object will be searched from different web pages in different web sites according to multiple dimension conditions 4. This search system executes calculation among a group search results according to conditions and generate combined result from a group search results, and return combined result according to conditions 5. This search system returns a list of combined results according to conditions Below are the scenario that the search system can be used for: 1. Search lowest combined price of a group products nearby a location, and the cost involve the transpiration 2. Search lowest combined price of a group products, which are sell in the same website, and sell by the same vendor, and the cost involve the logistics 3. Search lowest combined price of travel routine, include the transportation to target city, the price of the hotel, and the price of travel routine, all of such price information can be located in the same website or in the different website.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 27% of the total text.

Page 01 of 10

A cognitive search system to generate combined results by multiple dimension search

In this solution, there is a term of the major consumption, which is the first consumer in keywords want to buy, and all other consumption are the consumption the consumer want to buy when the consumer buy the major consumption, they might be optional consumption for the consumer. Currently, search engine can only search information for one target object, this solution built with in a new system, and a new format of keywords must be defined, and the system can provide a friendly input UI (User Interface), and can parse and learn the input keywords combination, and then execute search action one by one, and finally return expected merged results, in order to meet this requirement, A new format of keywords must be defined:

1.The keywords must define the relationship of all objects in one search,

for example: this is combined location-based price search

2.The keywords must define all objects, for example: all consumption

3.The keywords must define where the start-location of the consumer is and what kind of transportation if transportation cost is considered.

4.The keywords must define what is nearby means, for example in 2 km (kilometer) nearby the major consumption


Below is one example of keywords combination:

price((the first object)

+(nearby: 2km)

+( transportation: place-A)

+(~the 2nd object)

+(~the 3rd object)


Page 02 of 10


price() : means this is for price combination search

nearby: means that this is a location-based search

transportation: means that this search also involve other cost like transportation

() : each target object must be put into "()", and the major object must be put into the first(), and if the cost of transportation involves, then a start-location item must be defined.

+: afterwards object must be combined by "+"

~: means that need to found related favorite for this keyword

Search System

This new system is consists of four sub-systems: Keywords input and parsing sub-system, Search and analysis sub-system, Cache sub-system, Search output sub-system, as below picture:


Page 03 of 10


Page 04 of 10

Keywords input and parsing sub-system: Provide a friendly webUI (web User Interface) to operator for keywords input and parsing and learning the input keywords, save related favorite into "cache sub-system", and find related favorite from "cache sub-system" to search and then trigger "Search and analysis sub-system".

Search and analysis sub-system: This is a core sub-system, according to learnt keywords, this sub-system can trigger to search. This sub-system executes search against existing internet public API(Application Program Interface) of well-known 3


consumption:Restaurant A-set-meal

Note: The keywords input and parsing sub-system might find a related favorite for 2

                                              party websites, the search can also be executed on map service system, the search can also be executed on group-buying website. All search...