Parte 1 Flashcards
Define “Content Understanding”
The ability to find any arbitrary content leveraging keywords and matching on attributes
The ability to find any arbitrary content leveraging keywords and matching on attributes
Define “Content Understanding”
Define “User Understanding”
The ability to understand each user’s specific preferences and leverage those to return more personalized results
The ability to understand each user’s specific preferences and leverage those to return more personalized results
Define “User Understanding”
Define “Domain Understanding”
The ability to interpret words, phrases, concepts, entities, and nuanced interpretations and relationships between each of these within you own domain-specific context
The ability to interpret words, phrases, concepts, entities, and nuanced interpretations and relationships between each of these within you own domain-specific context
Define “Domain Understanding”
What are the three dimensions of user intent?
- Content Understanding
- User Understanding
- Domain Understanding
- Content Understanding
- User Understanding
- Domain Understanding
What are the three dimensions of user intent?
Summarize the end goal of AI-powered search
The optimal understanding of users and their query intent
The optimal understanding of users and their query intent
The end goal of AI-powered search
- Basic Keyword Search
- Taxonomies / Entity Extraction
- Query Intent
- Automated Relevance Tuning
- Self Learning
Search Intelligence Progression steps
Search Intelligence Progression steps
- Basic Keyword Search
- Taxonomies / Entity Extraction
- Query Intent
- Automated Relevance Tuning
- Self Learning
Basic Keyword Search: Step in Search Intelligence Progression steps and characteristics
Step 1
- Inverted index
- TF-IDF
- BM25
- Multiligual text analysis
- Query formulation
- Inverted index
- TF-IDF
- BM25
- Multiligual text analysis
- Query formulation
Basic Keyword Search: Step 1 of Search Intelligence Progression steps
Taxonomies / Entity Extraction: Step in Search Intelligence Progression steps and characteristics
Step 2
- Entity Recognition
- Taxonomies
- Ontologies
- Business Rules
- Synonyms
- Entity Recognition
- Taxonomies
- Ontologies
- Business Rules
- Synonyms
Taxonomies / Entity Extraction: Step 2 of Search Intelligence Progression steps
Query Intent: Step in Search Intelligence Progression steps and characteristics
Step 3
- Query Classification
- Semantic Query Parsing
- Semantic Knowledge Graphs
- Concept Expansion
- Automatic Query Rewrites
- Clustering
- Classification
- Personalization
- Question/Answer Systems
- Virtual Assistants
- Query Classification
- Semantic Query Parsing
- Semantic Knowledge Graphs
- Concept Expansion
- Automatic Query Rewrites
- Clustering
- Classification
- Personalization
- Question/Answer Systems
- Virtual Assistants
Query Intent: Step 3 in Search Intelligence Progression steps
Automated Relevancy Tuning: Step in Search Intelligence Progression steps and characteristics
Step 4
- Signal Boosting
- Collaborative Filtering
- AB Testing / Multi-armed bandits / Backtesting
- Genetic Algorithms
- Learning to Rank
- Signal Boosting
- Collaborative Filtering
- AB Testing / Multi-armed bandits / Backtesting
- Genetic Algorithms
- Learning to Rank
Automated Relevancy Tuning: Step 4 in Search Intelligence Progression steps
Self-learning: Step in Search Intelligence Progression steps
Step 5, final step
Step 5 in Search Intelligence Progression steps
Self-learning: Final step
Reflected Intelligence
Leveraging continual feedback loops of user inputs, content updates, and user interactions with content in order to continually learn and improve the quality of your search application.
Leveraging continual feedback loops of user inputs, content updates, and user interactions with content in order to continually learn and improve the quality of your search application.
Reflected Intelligence
What are the two extreme ends of a continuous personalization
spectrum within information retrieval
Search and Recommendation
What spectrum are Search and Recommendation the extremes of?
The two extreme ends of a continuous personalization
spectrum within information retrieval