Dynamic Dictionary Encoding With Position-Based Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional encoding methods using dynamic dictionaries result in redundant data when encoding data is reused, as the entire dynamic dictionary is stored in a compressed file, including data not related to the extracted code.

Innovation Solution

The proposed method involves using a dynamic dictionary with segments associated with position information, allowing for efficient dynamic code encoding by separating words and dynamic codes based on their positions, and creating an extraction dynamic dictionary that only includes relevant data, thereby preventing redundancy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the entire dynamic dictionary is stored in the compressed file to enable encoding reuse, then encoding data can be reused without reprocessing, but redundant data is included that is not related to the extracted code

Engineering Contradiction:

Improveencoding data reuse efficiencyVSAvoidredundant data storage

Core Design Contradiction:

ProductivityVSLoss of substance

Solution Approach 1:

The dynamic dictionary is divided into multiple segments based on position information in the text data. Each segment corresponds to a specific position range, allowing only the relevant segment to be stored in the compressed file rather than the entire dictionary. This segmentation enables efficient encoding reuse while eliminating redundant data from unrelated positions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Only the necessary portion of the dynamic dictionary (the segment corresponding to the extracted code position) is extracted and stored in the compressed file. The irrelevant portions are excluded, thereby removing redundant data while preserving the functionality needed for encoding reuse.

Inventive Principle:

Principle #2Taking out (Extraction)

2Loss of substance

If position information is associated with each code in the dynamic dictionary to enable selective extraction, then redundant data can be eliminated, but the complexity of the dictionary structure increases

Engineering Contradiction:

Improveredundant data storageVSAvoiddictionary structure complexity

Core Design Contradiction:

Loss of substanceVSDevice complexity

Solution Approach 1:

Position information is pre-associated with each code in the dynamic dictionary during the encoding process. This preliminary action enables subsequent selective extraction of only the necessary dictionary segments without requiring complex runtime analysis, thereby reducing overall system complexity despite the initial structuring overhead.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10404275B2Non-transitory computer readable recording medium, encoding method, creating method, encoding device, and decoding device

Publication Date: 2019.09.03 FUJITSU LTD

AI summary

An information processing apparatus encodes text data by using a dynamic dictionary in which codes and words appearing multiple times are associated with the text data. A registration destination area, in the dynamic dictionary, of each of the codes registered in the dynamic dictionary at the encoding is associated with a position, in the text data, of each of the words associated with the codes.