The core of the technology is a patented (in the US and Israel) method for integrating statistical algorithms with ontological based input, to imitate processes of human-reading of texts.
The combined knowledge of seasoned subject matter experts regarding texts of the same domain used in training are applied to the processed texts through algorithms. This method is called “Artificial Intuition”.
A human reaches “intuitive” conclusions – even by superficial reading – regarding the authorship and intent of a given text, subconsciously deriving them from previous experiences with similar texts or from linguistic knowledge relevant to the text. Then as the human accumulates more information through other features (statements, spelling, and references) in the text, he either strengthens his confidence in the initial interpretation or changes it.
The Meaning Mining technology extracts such implicit meaning from a text or the hermeneutics of the text. It employs the relationship between lexical instances in the text and ontology. *
Stages in the Meaning Mining process
Language and dialect detection
Our software identifies 80 languages and is capable of distinguishing between different languages using the same character sets. In addition, Meaning Mining can identify multi-lingual texts and hybrid language texts, such as “Spanglish” and “Frarabic”.
First Tier Domain and Topic Categorisation
Meaning Mining employs statistical algorithms to identify whether or not the input text is at all relevant to the topic that interests the user. This initial triage allows the user to focus on the issues that really are important without the need to invent ambiguous “keywords” and crawling rules.
Natural Language Processing
Meaning Mining Natural Language Processing performs morphological analysis, base phrase identification, part of speech analysis, syntactic analysis and engines for concatenation of lexical units in the text to larger meaningful units.
Named Entity Recognition, Analysis, Aggregation and Matching
Meaning Mining recognises analyses and aggregates entities (persons, places, organisations, events etc.) through application of statistical algorithms combined with semantic information. We can extract implicit and contextual information, such as: ethnicity, gender, affiliations, nicknames…
Inference of Implicit Ideas, Relations and Sentiment
Meaning Mining uses statistical and rule-base algorithms that combine information from the NLP engine, the knowledge base and the context to extract implicit information from the text.
Meaning Mining identifies the attitude of the author of the given text towards different entities, drills down into sentiment towards specific attributes or features of the entities and aggregates sentiment towards a set of linked entities with common “parent” entities. The sentiment analysis algorithms are based on statistical models composed of semantic features.
Meaning Mining identifies contextual information from the text regarding the relationships between entities mentioned. This includes: workplace relationships, family relationships, relationships between people and locations such as meetings, place of birth, place of residence etc.
Idea mining & Ontology-based Categorisation
Meaning Mining extracts “ideas” from the text. These “ideas” are specific concepts that do not represent a named entity but an action or event that may be important for the user (f.e. suicide bombing preparation). Meaning Mining categorises each document according to clusters of categories: type of document, priority (risk) level, ideological affiliation and domain. In addition, identifies topics within the domain.
Generation of reports
Meaning Mining generates a natural language report for any given document. The report characterises the document and presents the key elements in the document: categorisation, topics, priority, ideas expressed, sentiment, trends, etc – in their appropriate context.
Meaning Mining provides the ability to search the database to find combinations of entities or pieces of information.