Indice Score-Card Algorithm (ISCA):
This Algorithm is incredibly versatile. It can be used as a short-hand score card Predictor for Binary Target Classes (Two possible outcomes), it can be used as a Validation Metric in Classification Models and it can even be used as a Base Drafter in Automated Drafting and Document Assembly. ISCA was built specifically as a Legal Predictor, however in theory its use could be extended to one of a General Score Card Predictor in many other fields. There are four variations of ISCA: “GT1”, “GT2”, GT3” and “Gt4”. The “GT” stands for “Genetic Transmutation”, it was so named because it transmutes the make-up (“genetic print”) of Permutations into weight classes, (Indices). One of the advantages of ISCA is that unlike Machine Learning Algorithms it does not require historical Data to make its Predictions; it therefore becomes a leaner and more portable alternative. ISCA can however be used with a Base Machine Learning Classifier, the classifier would feed data to ISCA and it would employ its own calculations for its predictions. In most instances this is the preferred method when it comes to complex, large-scale Predictive Analytics projects.
For the purposes of this Demo, ISCA is purely for demonstrative purposes. Its features are not covered in this demo tutorial, nor is an explanation given regarding its outputs. ISCA requires fairly advanced knowledge of Lex Quant Summa, the LQS Manual provides detailed information on ISCA and is sufficient for the advanced use of ISCA. In addition, Quantum Data Works Workshops will also enhance the user experience and enable users to get the most out of ISCA.
An “Iteration” within the context of ISCA is the introduction of a single Event or Fact to the Prediction or Diagnostic enquiry, for example if ISCA predicted or diagnosed an outcome based on Five Iterations, the cumulative effect of Five Events/Facts lead to the final Prediction or Diagnosis.
ISCA’s Prediction enquiries, outputs and results are commensurate to the iterative nature of ISCA, that is to say each time a new Fact or Event is introduced to the Algorithm, ISCA presents its calculations to the user. This method allows for easier diagnosis of outcomes because it is possible to “pause” all events up to a desired point and extract the cumulative Data of preceding iterations. In very simple terms, it is possible to approximate which party was Losing/Winning, why and at which point.
ISCA is delineated into two core functions: the “Retro Diagnostics” and “Predict Verdict” functions. The former Diagnoses the Verdicts of decided Cases using the Iteration methodology, the latter is used as a Binary Target Class Predictor. Both of these processes have key Data outputs and can both occupy a single Workflow.
These Data outputs are an advantageous by-product of ISCA, they are the equivalent of Meta-Data that Predictions and the Diagnostics of Outcomes invariably produce. This Data is incredibly useful and LQS provides functions for the modelling of this Data to enable the extraction of key Insights and Analytics.
Activating ISCA’S Quantum Storage is an advanced feature of ISCA. As already discussed, Quantum Data is the cumulative mass of all the Data specific to a particular process, in this instance, ISCA. ISCA’s Quantum Storage consists of all Iterations, Data, Diagnostics and Results of previous ISCA executions. Like other tools that have Quant