Graph and Text Analytics
Graphwise Graph and Text Analytics comes with the PoolParty Extractor Service and Metadata Studio, both intelligent text mining tools based on semantic technologies. They combine machine learning and natural language processing techniques with knowledge graphs to enable algorithms to better analyze text by not only processing words, but understanding the underlying concepts and their context.
Why does this matter? Because unstructured data (a.k.a text) contains a lot of ambiguities that a machine cannot process unless told how to interpret it. PoolParty does this with little training data and automates tagging workflows so that you can have enriched metadata rather than a jumble of text fragments.
Extract meaning instead of strings
Graphwise Natural Language Processing extract meaningful phrases from a document to identify what content is about. The result is that the metadata is richer and more precise, helping to create a knowledge representation that accurately fit the company’s domain.
Automatically extract relevant concepts and specialized entities (e.g. product codes or file numbers), timestamps, names, or locations
Use machine learning based Named Entity Recognition (NER) capabilities to extract organizations, people, and locations
Uncover “masked” shadow concepts that are not explicitly written in the text
Detect the “aboutness” of a text using domain-specific sentiment analysis and machine learning algorithms (SVM, Deep Learning, Naive Bayes, and more)
Leverage corpus learning to denote value to a term based on its relevance to the text instead of frequency (our competitors go for the less meaningful approach)
Tackle linguistic challenges in your text with natural language processing.
Content is made up of natural language which can be tricky to interpret because it contains issues such as ambiguity in the same word meaning a different thing, i.e. apple like the fruit and Apple like the tech company. Graphwise Natural Language Processing doesn’t require complex rules to overcome these challenges.
Use word sense disambiguation to distinguish ambiguous terms based on local context of the term in a knowledge graph
Analyze words using lemmatization techniques based on a dictionary (available in multiple languages)
Extract the metadata which makes your content intelligent.
Text mining in a thesaurus serves as the backbone for semantic concept tagging. The benefit to maintaining tags in a taxonomy is the consistency it provides through its hierarchical structure and controlled vocabularies.
Make use of synonyms through preferred labels, alternative labels, and hidden labels
Sort concepts into their corresponding classes and concept schemes in the taxonomy for better categorization
Harmonize annotations with RDF data for graph databases in PoolParty Unified Views
Use the enriched metadata to benefit from a semantic search engine
Our automated concept tagging serves as the basis of ready-made integrations like Graphwise for M365, Tridion Sites, and Adobe Experience Manager. Use Graphwise to take your CMS to the next level.
Other features you’ll love:
Corpus Management
Improve the extraction of terms by relating them to a knowledge graph built on a set of documents representative of the domain, called a corpus.
Extraction workbench
With the Extractor’s test extraction function, you can configure and reproduce in a real-time GUI how the extractor works and can be optimized.
Multiple Corpora
You can extend the text-mining capabilities of Graphwise Natural Language Processing by uploading multiple corpora.
Semantic Classifier
Graphwise offers a unique tool to make classification of hundreds or thousands of digital documents in a breeze.
Graphwise Graph and Text Analytics can pull relevant entities from a text and tag thousands of documents in less than an hour.
Users can start right away:
You don’t need to be an IT person to use the PoolParty suite of tools. Anyone from your team can get quickly acquainted with your software.
“Initially, it took the engineering team up to two weeks to edit basic terms or single tags. Because of Graphwise, however, this time window has shrunk to less than 24 hours!”
