Location
- Morning tea
- Lunch
- Afternoon tea
July 27
- Introduction to text analytics
- Use cases and case studies (government and commercial)
- Building blocks of text analytics: language Identification, morphological analysis, entity extraction and entity linking
- Adapting Rosette for accuracy and performance
- Advanced Rosette capabilities: Sentiment analysis, categorisation and relationship extraction
Advanced use of Rosette Tools
Meeting Accuracy Goals
- Long and short string models for language identification
- Adapting entity extraction with lists and patterns
- Adapting entity extraction with machine learning
- Building custom categorizers / classifiers with keyword and machine learning methods
Meeting Performance Goals
- Distributed processing using SDK libraries (Hadoop/ example use case?)
- Scaling using our on-premise API
Integration options
- SDK
- API
July 28
Name and Identity Resolution
- Introduction to translating names and name matching
- Use cases and case studies
- Fuzzy name matching using Rosette
- Name matching at scale using Rosette with Elasticsearch and Solr
- Putting it all together:
Name matching, entity linking, relationship extraction and entity resolution
July 29
Personal Consultation:
One hour morning consultation with a Basis Technology senior customer engineer for “off syllabus” items.
- Brainstorm the challenges that are 'specific' to your team and project
Prerequisites
This course is aimed at Software Developers and Integrators. Attendees should have a working knowledge of the following:
- Search Indexing and Querying
- RESTful web services
- Command-line tools
- Scripting languages: Perl / Python / Ruby
- Linux file system navigation and environment variables
There is no requirement to have tried using any particular software before attending the course. However it may be beneficial for attendees to have installed and become familiar with the following:
- Elasticsearch 1.75+
- Solr 4.6+
- Python 2.7+
- Ruby 1.8+
- cURL
- Postman for Chrome (or another GUI based tool for interacting with RESTful web services)
- A free trial API key from:
https://developer.rosette.com/signup
Instructor: Declan Trezise
Declan Trezise is the Senior International Customer Engineer for Basis Technology and based in the UK.
Declan has many years of experience working with challenging problems; converting unstructured text into actionable insight, communicating complex technical solutions to audiences at varying levels and walks of life. Declan was also the face of cybersecurity and data analytics for BAE Systems Applied Intelligence (Detica) as the manager and presenter at the renowned “NerveCentre” facility in the London.
In a former role at Monster.com’s Government Solutions division Declan was part responsible for the delivery of the UK’s largest job board, Universal Jobmatch for the Department for Work and Pensions. He trained as a theoretical physicist at the University of Exeter and as a graduate worked with a range of technologies in intelligence applications from network probes for lawful interception to data retention and querying systems for the largest telecommunications companies.