Knowledge Discovery
Go beyond standard search and recommendation with technology that understands user intent and context to retrieve the most relevant results
Making the knowledge and data of an organization available in its daily work is just as much a basic function of a successful company as its energy supply or human resources. Today, search and recommendation tools are a business standard; if you want to survive in the market with your knowledge-intensive products and stand out from the competition, you have to invest in knowledge discovery that makes the difference.
Graph Search
Graph search delivers results even in those cases where standard search engines deliver only imprecise results or even fail. Semantic search incorporates both user intent and the contextual meaning of terms into the query. On the one hand, this allows for more accurate results, while on the other hand, it also allows for intervention in search facets and search depth.
Graph search finds results that match the meaning of a user’s query, rather than focusing solely on exact words and phrases. For example, when a user enters the term “Apple,” semantic search recognizes from the context that the user is talking about an electronic device, not a fruit, and because it also uses concepts instead of simple terms, it returns related results for, say, “Mac” – something that is nearly impossible with a pure keyword search.
Graph Recommender
In a company, recommendation systems can only be helpful if they are tailored to specific use cases, fit the requirements of employees, and comply with industry standards. The technological approach of a recommender system is therefore crucial. We believe that knowledge-based recommender systems offer the best value for the most appropriate corporate set of features.
Collaborative Filtering
Content Based
Knowledge Based
Specific to Domain Knowledge
GraphDB Free
Cold Start Capable
GraphDB Enterprise
User Intent Driven
GraphDB Enterprise
Independent of Early Raters
GraphDB Enterprise
Open to Knowledge Engineering
GraphDB Enterprise
Insensivity to Rating Sparsity
GraphDB Enterprise
Insensivity to Preference Changes
GraphDB Enterprise
Graph search finds results that match the meaning of a user’s query, rather than focusing solely on exact words and phrases. For example, when a user enters the term “Apple,” semantic search recognizes from the context that the user is talking about an electronic device, not a fruit, and because it also uses concepts instead of simple terms, it returns related results for, say, “Mac” – something that is nearly impossible with a pure keyword search.
Features that boost your corporate knowledge retrieval
Altogether, our graph capabilities allow you to transform inconsistent metadata often found in siloed systems into interoperable knowledge graphs that build the foundation for intelligent applications. These applications use your data more effectively so that you can benefit from a more productive workforce and a happier customer base.
Unlocking Data
Harness the data that other engines leave out because of missing metadata, siloed storage, inappropriate structure, or wrong format.
Active Exploration
Queries are not static. They can be expanded, narrowed, refocused and recontextualized. You get control to navigate to the content you are looking for.
Polishing the Metadata
Incorrect or outdated metadata is semantically corrected, enriched, and mapped to make the corresponding content available for search again.
Adapt to the user
Different people ask about the same thing differently. Adapt to the user’s jargon so that the search engine speaks the user’s language and not the other way around.
Arguments that justify your investment
Relevant results in less time
According to IDC, a knowledge worker spends about 30% of their workday searching for information – mainly sifting through long lists of search results. As we see with our customers, this effort can be halved if the results are narrowed down by user and context and the search results are better navigated. So we’re talking about a 10-15% efficiency boost from Graph Search.
Enable employees
It is a dilemma! You need to know the terminology of a domain to be able to make successful queries. Graph search puts an end to this, because it translates your query into the jargon of the domain and thus finds content according to meaning and not according to word similarity. This makes knowledge accessible to people outside the domain and enables collaboration between novices and experts.
Data-grounded decision-making
Acting focused and on the basis of a solid knowledge base is key in any decission making – form design to business. Filtering out the noice and keep the access to their own trusted sources is key. Graph Search delivers actionable data for a solid data-grounded picture of all business objects and processes.
Value to legacy content
After a certain lifetime, old content can become difficult to be used because data structures, conventions and formats have changed over time and informal knowledge about it also leaves the company together with colleagues. Semantic knowledhe retrieval keeps this content connected and thus generates added value again and again, for example, by ensuring that the service department still has older content at hand.
Built for you!
Out-of-the-box solutions quickly reach their limits and it can happen that your recommendation system does not generate the added value you expect. The Graphwise konwledge retrieval systems are as flexible as your company’s needs are. We have been integrating knowledge systems for companies for over 20 years, so our systems are prepared for your specific circumstances by design.
Deep Dive in Knowledge Retrieval Leveraging Semantic Footprinting
In essence, the graph-based approach to knowledge retrieval assumes that each knowledge object can be given a unique position in the knowledge graph. We call this semantic footprinting. This position represents the combination of inherent content description and contextualization within the knowledge space. This combination is the core element of graph-based search and recommendation systems, which maps usable relations between content objects independent of content type and content usage (user behavior).
While other systems go through untraceable processes, knowledge-based recommendation systems follow a clearly defined and step-by-step explainable and transparent logic. The user submits the query (1), which consists of a sentence, a paragraph, a section, or an entire document. The text passes through the text annotation component (2) and receives its semantic footprint (3). The subsequent query expansion (4) is a traceable intervention in recommendation depth and recommendation sharpness. The matches found (5) thus contain both the implicit knowledge of the domain model and the query-specific adjustments. The finally obtained retrieval (6) is clearly traceable and controlled.
Thus, the Graphwise approach for knowledge retrieval provides us with high domain specificity and adaptability to the particular application domains. It delivers relevant results since it has no blind spots in filtering behavior. Moreover, with the adaptable architecture that Graphwise search and recommendation systems offer, we can combine knowledge graph entities with statistical AI implementations (ML, LLMs), thus combining capabilities from both worlds.
Useful Resources
Webinar
Agile Taxonomy Management for Customer Satisfaction.
Healthcare Information System to Australian Citizens using PoolParty Semantic Suite
DEMO
Try out
Download the White Paper
This white paper gives an overview on costs & benefits of implementing a Knowledge Graph