Scoring and Profiling



Scoring and profiling are key processes are key to providing valuable insights concerning a subject (such as an individual or asset) from large sets of data. Data scoring and profiling and produce critical insights concerning target cohorts and can be tailored to meet specific requirements and goals.

Scoring allows the metrics of key risk and vulnerability indicators to be reflected in a score, allowing an at-a-glance rating to help assist organisations in prioritising the provision of care and support. Scoring can be configured to meet key programme aims and objectives. Within scoring, multiple weightings can respond to changes in status and reflect different severities in terms of context, recency, and event.

Profiling adds a layer of qualitative interpretations on top of multiple indicators. It allows metrics and scoring to contribute to a risk and vulnerability profile based on past analysis and trends. Profiling is a powerful method of building a data-based context of a subject, it enables earlier and more effective intervention strategies. Scoring and Profiling play a foundational role in predicative analysis.  

What is it?

Scoring and Profiling are the uses of an integrated data set to provide valuable insights around risk, opportunity, and trends.

Scoring

Scoring is at the root of assigning a numerical score to a field value. It involves the collection and generation of useful metrics and metadata from a numerical analysis of people, groups, trends, assets, products.

Any number of scoring rules can be set up to act on any number of fields.

Scoring rules can be combined into categories (e.g. health, education, police, different asset dbs, customer transactions)

The Score is retained as a snapshot against an individual, group or asset and is run on a scheduled basis to produce a chronological view of trends and or patterns. Allows us to see how things are changing over time.

Profiling

Subject Profiling is the qualitative interpretation of these values that allows insights to be gained, discovering and understanding the relationships, links, or risks associated with numerical markers of scoring. Thresholds can be set and compared to national standards, averages or past cases.

Types of Profiling:

  • Qualification: the establishment of criteria for subjects to meet a required threshold. Qualifying criteria can be tailored to meet the needs of what needs to be differentiated. A basic example might be types of absence: is a child repeatedly absent from education or long-term absent? Qualifying profiling allows the control of determining the criteria of a subject to qualify categorisation
  • Prioritisation: a type of profiling that builds urgency and risk into a view. It is a way of determining which subject should be prioritised, a way of understanding who or what is most in need of immediate action
  • Categorisation: allows the management of determining factors, if a subject falls within a particular category

Outcomes and Benefits

  • No automated profiling or decision making, you control the results of profiling
  • Allows informed decisions to be made by professionals with the right expertise and qualitative case knowledge
  • Allows organisation to be pro-active and data-driven, supporting intervention, early action and more effective support service provision
  • Profiling of a unified data set from disparate data sets can assist in the mitigation of poorly integrated or siloed data. Building a corporate data asset

The Sentinel Data Platform provides the following Profiling and Scoring Insight Capabilities and Features

  • Multiple scores to one: repeated trigger does not increase score or record frequency
  • Multiple scores to many: useful for households, associations
  • Set complexity
  • Conditional scoring: such as age, risk, red flags, special needs, histories and relationships
  • Fuzzy searches on data
  • Profiling and scoring harmonized with Golden Records
  • Identify shared characteristics across individuals to place them in demographics/families
  • Harmonise users across systems using the sentinel key to allow for use of all data in profiling
  • Insights from all of data quickly
  • Pattern prediction
  • Risk Analytics
  • Identify demographics and families
megan-wu69n8UPhP4-unsplash

Liverpool City Council

Helping one of the UK's largest local authorities focus frontline services identify individuals and families who were most vulnerable. We provided an up-to-date, accurate single view of data, integrating with over 40 different data sources - including 4 from the NHS.

Read their story

iStock-1149461162

Devon

Using the Sentinel Supporting Families Solution, Devon County Council were able to more effectively identify families with multiple complex needs and dependencies.

The quality of data available across the authority and its partner agencies has been enhanced and is far more visible.

Read the story

iStock-514584455

Birmingham City Council

Adopting our Early Years Funding Portal means that local authorities as large as Birmingham can make light work of the high volume of child records and eligibility checks that need to be processed.

Read their story

Profiling

Profiling individuals or assets can provide a score, a fast and clear verdict for which subjects need to be prioritised for action. Actions can be automated and triggered by certain scores, allowing more effective targeting of resources.

All profiling is configurable to suit your goals or programme requirements, allowing users to quickly and easily see what is of greatest priority. 

Scoring

Using a set of configurable rules, scores can be assigned against a person, property or asset. These scores can reflect an overall status such as crossing an eligibility threshold or time for actions such as intervention or servicing or they can operate on different levels need and requirement.

Scores can also be weighted, so certain changes have more impact on priority than less urgent changes. Scores can also reflect recency, for example scores from historic events can, if necessary, be less heavily weighted than scores from very recent events.  

Scoring Rules

Because scoring rules are configurable, you can tailor them to meet your changing needs and requirements. If programme requirements or company policy change you can update the scoring rules accordingly.

Clauses can be added to identify which individual or asset fall under a particular project phase and score according to phase. This flexibility and control allow scoring to reflect your needs and serve as an effective signal of urgencies and priorities.  

Transform data use in your organisation

Book your free data assessment today, and find out how much of an impact Sentinel's Master Data Management tools can have on your business.

  • Strict control and monitoring of data quality
    and completeness
  • Built using the ICO's "data protection by design"
    approach
  • Trusted by public sector organisations and local authorities
  • Experienced, dedicated team of data integration and data sharing specialists

Take the hassle out of data management. Call us on +44(0)800 612 2116 or email us [email protected].