System and process for natural language processing and reporting

• US 9,619,555 B2
• Filed: 10/02/2014
• Issued: 04/11/2017
• Est. Priority Date: 10/02/2014
• Status: Active Grant

• Current Assignee: Shahbaz Anwar
• Original Assignee: Shahbaz Anwar
• Inventors: Anwar, Shahbaz, Malik, Muhammad Kamran, Javed, Muhammad, Abbas, Faheem, Saeed, Imran
• Primary Examiner(s): MOSER, BRUCE M
• Application Number:
  US14/505,323
• Publication Number:
  US 20160098479A1
• Time in Patent Office: 922 Days
• Field of Search: None
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 40/205 Parsing

Abstract

The present invention is directed to a computerized system and process for natural language query and reporting comprising a processor, memory, and a query interface configured for receipt of a data source selection. The processor searches and retrieves over a network unstructured data based on the received data source selection, parses the unstructured data into data blocks and stores the data blocks in a local database. The processor semantically parses the data blocks and stores the resulting data in a structured database.

A report module is configured to iteratively receive keyword input and instantiate a subject node, the node representing a subset of data blocks of the structured database having the input keywords. The report module creates a taxonomy based on the input keywords, with descendant levels representing a subset of data blocks of the subject node, the subset of data blocks having the input keywords combined with other words in the data blocks. The report module further associates a keyword selector with the subject node, the keyword selector presenting an interface for additional user keyword input.

Claims

1. A system for natural language query and reporting comprising:

a processor and memory;

a query interface configured for receipt of a data source selection;

said processor configured to search and retrieve remote over a network unstructured data based on the received data source selection, parse unstructured data into data blocks and store said data blocks in a local database;

said processor configured to semantically parse said data blocks, including part of speech tagging, and store resulting data in a structured database;

a report module configured to;

receive a natural language query;

semantically parse said natural language query, including part of speech tagging;

order and rank by comparison of keywords in the natural language query with words in the data blocks of the structured data;

assigning an association value among keywords within said natural query;

creating a taxonomy of keywords of said natural language query, where the keyword groups of above a threshold association value are assigned a parent child relationship;

creating a node family of the same structure as said taxonomy, each of said nodes representing a subset of data blocks of said structured database having the respective keyword;

associating a keyword selector with said subject node, said keyword selector presenting an interface for additional user keyword input;

associate a report selector with said subject node, said report selector configured to receive a report type and a node and generate a report based on data blocks associated with said node.

2. The system of claim 1, wherein said processor is further configured to assign sentiment values to said data blocks.

3. The system of claim 1, wherein said keyword selectors comprises a lists of words of a single part of speech, said part of speech selected from nouns, verbs, and adjectives.

4. The system of claim 1, wherein said report types include are selected from the following:

bar chart, order squares chart, matrix chart, tree map chart, sankey chart, text chart.

5. The system of claim 1, wherein said keyword selector comprises a list of sentiment values.