Concept-based method and system for dynamically analyzing results from search engines
First Claim
1. A computer-readable storage medium comprising code representing instructions to cause a processor to:
- receive, from a user, a first selection indicating a predefined corpus of a plurality of objects, the plurality of objects being of a finite number;
extract, from the plurality of objects, a plurality of all concepts that occur in the corpus, each concept from the plurality of all concepts being one or more of;
a part of speech;
a geographic region;
a person;
a group of persons;
a business unit;
an organization;
an action;
a process;
a product;
a service;
a subject;
ora category;
receive, from the user, a second selection indicating a first concept and a second concept from the plurality of all concepts, wherein the first concept includes a seed concept and at least one related concept, and wherein the second concept includes a seed concept and at least one related concept;
calculate a strength of a relationship between the first user-selected concept and the second user-selected concept, the relationship being based at least in part on a presence of both the first user-selected concept and the second user-selected concept in each of at least a first object and a second object in the corpus, the calculating including receiving, from a user, a proximity value for the first user-selected concept and the second user-selected concept, the proximity value being based at least in part on a number of words occurring between the first user-selected concept and the second user-selected concept within each object from the plurality of objects; and
identify a third concept when the strength of the relationship exceeds a predetermined threshold, the third concept being the relationship between the first concept and the second concept, the predetermined threshold being based at least in part on a predetermined proximity threshold value, the predetermined proximity threshold value being based at least in part on a number of words between two concepts within an object.
8 Assignments
0 Petitions
Accused Products
Abstract
A method, system and computer program for identifying a relationship between concepts in an object includes (i) selecting the object for inclusion in an initial subset of objects obtained as an output of a search engine; (ii) storing the object in a database; (iii) creating an index of the object; (iv) analyzing the index to extract two concepts relating to the object; (v) highlighting the two concepts within object; (vi) calculating a strength of a relationship between the two concepts using a ratio; (vii) storing the two concepts and the strength of the relationships between the two concepts in a database; (viii) presenting the two concepts and the strength in an interface that has a concept matrix; and (ix) repeating steps vi through viii for every possible pairwise combination of concepts within the index. The relationship can be the two concepts both occur in the same object; the two concepts occur near each other within the same object; and the two concepts occur near each other and in a particular order within the same object. The ratio can be a support ratio; an occurrence ratio; and a combined ratio. An additional concept can be added to the index; all the concepts can be grouped in the index by concept type; and complex meta-structures can be created. The concept type can be a part of speech; a geographic region; a person; a group of persons; a business unit; an organization; an action; a process; a product; a service; a subject; and a category.
240 Citations
21 Claims
-
1. A computer-readable storage medium comprising code representing instructions to cause a processor to:
-
receive, from a user, a first selection indicating a predefined corpus of a plurality of objects, the plurality of objects being of a finite number; extract, from the plurality of objects, a plurality of all concepts that occur in the corpus, each concept from the plurality of all concepts being one or more of; a part of speech; a geographic region; a person; a group of persons; a business unit; an organization; an action; a process; a product; a service; a subject;
ora category; receive, from the user, a second selection indicating a first concept and a second concept from the plurality of all concepts, wherein the first concept includes a seed concept and at least one related concept, and wherein the second concept includes a seed concept and at least one related concept; calculate a strength of a relationship between the first user-selected concept and the second user-selected concept, the relationship being based at least in part on a presence of both the first user-selected concept and the second user-selected concept in each of at least a first object and a second object in the corpus, the calculating including receiving, from a user, a proximity value for the first user-selected concept and the second user-selected concept, the proximity value being based at least in part on a number of words occurring between the first user-selected concept and the second user-selected concept within each object from the plurality of objects; and identify a third concept when the strength of the relationship exceeds a predetermined threshold, the third concept being the relationship between the first concept and the second concept, the predetermined threshold being based at least in part on a predetermined proximity threshold value, the predetermined proximity threshold value being based at least in part on a number of words between two concepts within an object. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A computer-readable storage medium comprising code representing instructions to cause a processor to:
-
receive a first user selection, the first user selection indicating a subset of a plurality of objects, the subset comprising a predefined corpus, the predefined corpus having a plurality of concepts, each concept from the plurality of concepts being one or more of; a part of speech; a geographic region; a person; a group of persons; a business unit; an organization; an action; a process; a product; a service; a subject;
ora category; extract a plurality of concepts that occur in the corpus; receive a second user selection, the second user selection indicating a first user-selected concept and a second user-selected concept from the plurality of concepts, wherein the first user-selected concept includes a seed concept and at least one related concept, and wherein the second user-selected conceit includes a seed conceit and at least one related conceit; calculate a strength of a relationship between the first user-selected concept and the second user-selected concept, the relationship being based at least in part on; a presence of both the first user-selected concept and the second user-selected concept in each of at least a first object and a second object in the corpus; a proximity of the first user-selected concept to the second user-selected concept within at least the first object in the corpus, the proximity based on a user-selected proximity value; and a number of occurrences of the first user-selected concept and the second user-selected concept within the first object in the corpus; and identify a third concept when the strength of the relationship exceeds a predetermined threshold, the third concept being the relationship, the predetermined threshold being based at least in part on a predetermined proximity threshold value, the predetermined proximity threshold value specifying a number of words between two concepts within an object. - View Dependent Claims (19, 20, 21)
-
Specification