System and methods and user interface for searching documents based on conceptual association

• US 8,762,371 B1
• Filed: 01/27/2013
• Issued: 06/24/2014
• Est. Priority Date: 03/16/2009
• Status: Expired due to Fees

• Current Assignee: Linfo IP LLC (Pueblo Nuevo LLC)
• Original Assignee: Guangsheng Zhang
• Inventors: Zhang, Guangsheng
• Primary Examiner(s): MOSER, BRUCE M
• Application Number:
  US13/751,120
• Time in Patent Office: 513 Days
• US Class Current: 707/730
• CPC Class Codes: G06F 16/3346 using probabilistic model
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A system and methods and user interface are disclosed for searching documents based on conceptual association, and for ranking documents based on content characteristics. A computer processing system receives a query containing a word or phrase that is a name of an object or concept, and can also receive a value for a customized ranking parameter. A computer storage device stores a dataset related to the object or concept name, wherein the dataset includes a plurality of property names and can also include association-strength values. A computer processing system can count a first frequency of a first property name and count a second frequency of a second property name in a document containing text in a natural language, calculate a relevance score as a function of the first frequency and the second frequency, and rank the plurality of documents using their respective relevance scores, and return one or more documents to the user based on the ranking of the plurality of documents. The function is in part defined by the customized ranking parameter.

Claims

1. A system for searching documents and ranking search results based on conceptual association, comprising:

a computer;

configured to store one or more datasets each being related to an object or concept, and to store a plurality of documents each comprising one or more words or phrases, wherein each dataset comprises one or more words or phrases as names of properties associated with the object or concept,

to receive a query, wherein the query comprises a word or a phrase that represents an object or concept name,

to retrieve a dataset related to the object or concept name in the query, the dataset including a first property name and one or more second property names,

to count a first frequency of the first property name in one or more of the plurality of documents, and a second frequency of the one or more second property names in the one or more documents, and to calculate a relevance score for the one or more documents as a function of at least the first frequency and the second frequency, and

to output the one or more documents or their representations as a search result based on the relevance score, wherein each of the one or more documents contains at least one word or phrase that matches a property name in the dataset, but does not match the word or phrase in the query.

2. The system of claim 1, further comprising:

a user interface configured to accept a query containing one or more concept names, and to accept a user indication to initiate a concept-based search.

3. The system of claim 2, wherein the user interface is further configured to display a label indicating that the search can be based on concept or association.

4. The system of claim 2, wherein the user interface is further configured to accept a customized ranking parameter for ranking the search results based on the content characteristics of the documents.

5. The system of claim 4, wherein the types of the document content characteristics includes at least a general contents versus a specific content, a brief content versus a comprehensive content, and a keyword-focused content versus a conceptually-associated content,

wherein the general content or the specific content is defined based on whether the content contains top N property names from a list of property names ranked by the strength of association,

wherein the brief content or the comprehensive content is defined based on whether the document content contains more than a pre-defined amount of terms representing properties associated with the object or concept,

wherein the keyword-focused content or the conceptually-associated content is defined based on whether the content is retrieved based more on terms that match the object or concept name than on terms that do not match the object or concept name but represent properties associated with the object or concept.

6. The system of claim 4, wherein the computer is further configured to receive the customized ranking parameter, and to calculate the relevance score as a function of the first frequency, the second frequency, and the customized ranking parameter.

7. The system of claim 1, wherein the first property name or one of the one or more second property names in the dataset is further associated with a numerical value, wherein the numerical value represents the association strength between the property and the concept or object, wherein the function for calculating the relevance score is further based on the numerical value.

8. The system of claim 7, wherein the numerical value is above or below a predetermined threshold value.

9. The system of claim 7, wherein the function further depends on a first multiplication of the first frequency and its corresponding numerical value;

and a second multiplication of the second frequency and its corresponding numerical value.

10. The system of claim 1, wherein the first property name is the same as the object or concept name in the query, and the one or more second property names are different from the object or concept name in the query, wherein the computer is configured to calculate the relevance score further based at least on the relationships between the first frequency and the second frequency.

11. The system of claim 10, wherein the computer is further configured to sum the frequencies of property names that are different from the object or concept name in the document to produce a total count, and to calculate the relevance score using a ratio of the first frequency to the sum of the first frequency and the total count.

12. The system of claim 11, wherein the relevance score is used to determine the degree of the document content characteristics in terms of being a keyword-focused content in contrast to a conceptually associated content, wherein the computer is further configured to rank or re-rank the documents based on the characteristics, wherein the keyword-focused content or the conceptually-associated content is defined based on whether the content is retrieved based more on the terms that match the object or concept name than on the terms that do not match the object or concept name but represent properties associated with the object or concept.

13. The system of claim 1, wherein the computer is further configured to:

count the frequencies of the plurality of property names in the one or more of the plurality of documents;

sum the frequencies of property names to produce a total count in the one or more of the plurality of documents; and

calculate the relevance score for the one or more of the plurality of documents based on the total count.

14. The system of claim 13, wherein the frequencies of the plurality of property names is counted up to a predetermined upper bound, wherein the relevance score is calculated using a ratio of the total count to the predetermined upper bound, wherein the total count is set to be equal to the predetermined upper bound if the total frequencies of the plurality of property names exceed the predetermined upper bound.

15. The system of claim 13, wherein the relevance score is used to determine the degree of the document content characteristics in terms of being a brief content in contrast to a comprehensive or detailed content, wherein the computer is further configured to rank or re-rank the documents based on the characteristics, wherein a brief content or a comprehensive or detailed content is defined based on whether the document content contains more than a pre-defined amount of terms representing properties associated with the object or concept.

16. The system of claim 7, wherein the computer is further configured to separate the plurality of property names in the dataset into a first group and a second group, wherein the numerical value associated with each of the property names in the first group is at or above a predetermined value, wherein the numerical value associated with each of the property names in the second group is below the predetermined value, wherein the computer is further configured to count the frequencies of the property names in the first group and count the frequencies of the property names in the second group in the one or more of the plurality of documents, wherein the computer is further configured to calculate the relevance score as a function of the frequencies of the property names in the first group and the frequencies of property names in the second group.

17. The system of claim 16, wherein the relevance score is used to determine the degree of the document content characteristics of containing general information in contrast to containing specific information about the object or concept, wherein the computer is further configured to rank or re-rank the documents based on the characteristics, wherein the document content characteristics of containing general information or containing specific information is defined based on whether the document content contains top N property names from a list of property names ranked by the strength of association.

18. The system of claim 16, wherein the computer is configured to sum the frequencies of the one or more property names in the first group in the document to produce a first sum and to sum the frequencies of the one or more property names in the second group in the document to produce a second sum, where the relevance score is calculated based on the first sum and the second sum, wherein the relevance score is used to determine the document content characteristics of containing general information in contrast to containing specific information about the object or concept, wherein the computer is further configured to rank or re-rank the documents based on the characteristics, wherein the document content characteristics of containing general information or containing specific information is defined based on whether the document content contains top N property names from a list of property names ranked by the strength of association.

19. A computing system for searching documents, comprising:

a computing device;

configured to receive a plurality of documents from a storage, and a query comprising a keyword, and a first user indication for a concept-based search, and to perform the concept-based search by;

(a) receiving a query, wherein the query comprises a word or a phrase that represents an object or concept name,

(b) obtaining a dataset based on the name of the concept or object, wherein the dataset comprises one or more terms representing one or more properties associated with the concept or object,

(c) identifying a first term in the plurality of documents, wherein the first term does not match the keyword in the query, wherein the first term matches one of the one or more terms in the dataset, and

(d) returning at least one of the plurality of documents containing the first term as a search result.

20. The system of claim 19, wherein the computing device is further configured to provide a user interface object to indicate that the concept-based search is an available search option, wherein the interface object is associated with a label, wherein the label is a text representing the meaning of “

concept”

or “

association”

,

to receive a second user indication for a keyword-based search, and

to provide an adjustable scale to receive a user preference towards either the concept-based search or the keyword-based search, wherein the computing device is further configured to use the scale value as a searching or ranking parameter.