Semantic Direction Engine for Automated Prospect Targeting

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current Business Intelligence, CRM, and Web Analysis Systems rely on subjective human judgment for keyword selection, leading to inaccurate results due to linguistic issues and the need to sift through numerous keywords, resulting in mistargeting and failure to identify proper prospects.

Innovation Solution

A computer system and method using machine intelligence to define a 'Semantic Direction' for content by converting words into numerical representations, eliminating the need for keywords and leveraging word2vec AI models to analyze large datasets, providing clear and actionable insights on cross-sell, new sales, and larger sales opportunities by measuring signal strength and intent across business entities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If marketers manually select keywords to determine prospect search intent, then they can identify potential prospects, but the results are inaccurate due to linguistic issues and multiple word meanings

Engineering Contradiction:

Improveaccuracy of prospect identificationVSAvoiddifficulty of keyword selection

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent replaces the manual mechanical process of keyword selection with an automated computational system using natural language processing and machine learning algorithms. The system automatically analyzes content to extract semantic directions and identify prospects, eliminating the need for marketers to manually search through thousands of keywords and resolve linguistic ambiguities.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent transforms the approach from discrete keyword matching to continuous semantic space representation. By converting content into numerical vectors in a high-dimensional semantic space, the system captures nuanced meanings and relationships between concepts, allowing for more accurate intent determination beyond simple keyword presence.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If marketers search through 2,000 or more keywords to find product-related words, then they can cover more ground, but the setup time and complexity increase significantly

Engineering Contradiction:

Improvecoverage of product-related termsVSAvoidsetup time for keyword selection

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary automated analysis of product content and market data to pre-compute semantic directions and relationships before prospect identification begins. This preliminary processing creates a ready-to-use semantic framework that eliminates the need for marketers to manually search through extensive keyword lists during the actual prospecting process.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses word embedding techniques to create compressed numerical representations (vectors) that capture the essence of thousands of keywords and their relationships. Instead of manually evaluating each keyword, the system works with these compact vector representations that encode semantic information from large corpora, dramatically reducing setup time while maintaining comprehensive coverage.

Inventive Principle:

Principle #26Copying

3Device complexity

If single words are used for intent analysis, then the process is simple, but words with multiple meanings lead to inaccurate results

Engineering Contradiction:

Improvesimplicity of analysis methodVSAvoidaccuracy of intent determination

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent transitions from one-dimensional keyword matching to high-dimensional semantic space analysis. By representing words and content as vectors in multi-dimensional space where dimensions capture different aspects of meaning, the system can distinguish between different contexts and meanings of the same word based on its surrounding context and position in the semantic space.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system creates composite semantic representations by combining multiple features including word embeddings, term frequency, inverse document frequency, and contextual information from surrounding text. This composite approach synthesizes multiple signals to determine intent, making the analysis more robust to individual word ambiguities while maintaining manageable complexity through automated processing.

Inventive Principle:

Principle #40Composite materials

Data Source

PatentUS11238233B2Artificial intelligence engine for generating semantic directions for websites for automated entity targeting to mapped identities

Publication Date: 2022.02.01 THE DUN & BRADSTREET CORP

AI summary

A method and system for employing a Language Processing machine learning Artificial Intelligence engine to employ word embeddings and term frequency-inverse document frequency to create numerical representations of document meaning in a high-dimensional semantic space or an overall semantic direction. This semantic direction can be used to quantitatively measure semantic similarity between online content consumed by a potential prospect and a given product or product family. The AI can automate the process of creating audiences for on-line marketplaces for programmatic advertising purposes by using representative product descriptions, such as a grouping of product descriptions for scalable, cloud-based databases, and then creating a hyper-focused intent-based audience based on companies that are showing a significant increase in intent.