Actionable Phrase Linking for Knowledge Base Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for how-to guides and knowledge bases fail to effectively inter-link actionable phrases across documents, leading to inefficient troubleshooting and customer care, as they are often poorly organized and lack rigorous processes for content creation and optimization.
Innovation Solution
A method and system that identifies actionable phrases in documents, uses a scoring function to link them to relevant information in a document corpus, and creates hyperlinks for easy access, utilizing a processor and memory to analyze parts of speech and dependencies between verbs and objects, and a search component to rank relevant documents based on a scoring function.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If how-to guides are created independently by contributors without rigorous organization processes, then content creation is fast and flexible, but the knowledge base becomes poorly organized and difficult to navigate
Solution Approach 1:
The patent segments the knowledge base into individual actionable phrases (verbs or verb phrases) that can be independently identified and linked. Each actionable phrase is treated as a discrete unit that can be hyperlinked to relevant documents, allowing the system to maintain the flexibility of independent content creation while enabling organized access through these segmented units.
Solution Approach 2:
The patent introduces hyperlinks as an intermediary mechanism between actionable phrases and relevant knowledge base documents. This intermediary allows users to navigate from a specific actionable phrase to related instructional content without requiring the entire knowledge base to be rigidly structured, thus maintaining ease of operation while preserving content creation flexibility.
2Device complexity
If virtual agents provide pointers to single knowledge base entries, then the system is simple to implement, but troubleshooting sessions become longer and less efficient when users need additional information
Solution Approach 1:
The patent applies preliminary action by pre-creating hyperlinks from actionable phrases to relevant knowledge base documents during the content creation phase. This allows the virtual agent to immediately provide comprehensive troubleshooting information without requiring additional interactions, thus reducing troubleshooting session duration while maintaining relatively simple system implementation.
Solution Approach 2:
The patent merges multiple pieces of troubleshooting information into a single interactive experience by hyperlinking actionable phrases to relevant documents. This allows users to access multiple related instructions within one troubleshooting session rather than requiring separate interactions, effectively combining multiple information sources without significantly increasing system complexity.
3Device complexity
If existing systems do not inter-link actionable phrases across documents, then document structure is simple and easy to maintain, but relevant information cannot be efficiently accessed from other parts of the knowledge base
Solution Approach 1:
The patent adds a new dimension to the document structure by introducing hyperlinks that create connections across documents. This dimensional addition allows information to be accessed from multiple entry points without fundamentally complicating the underlying document structure, thus maintaining simplicity while improving information accessibility.
Data Source
AI summary
A computer-implemented linking system and method provide for linking actionable phrases in a first document to other documents in a document corpus. The method includes identifying at least one actionable phrase in a first document. The actionable phrase may include an action, its direct object, and any modifier of the direct object. For each identified action phrase the document corpus is searched to identify other documents, which are scored using a scoring function which takes into account occurrences of words of the actionable phrase in each identified document. The actionable phrase is linked to at least a part of one of the most highly ranked documents in the set of documents.