AIC 300

Combination of nodes

A representation of data attributes in a classification tree.

Motion to dismiss

A request that a court terminate an action because of settlement, voluntary withdrawal, or procedural defect.

Wearable sensor tag

A sensor attached to or embedded in clothing and accessories.

Federal Rules of Civil Procedure

A set of rules established to ensure that civil actions and procedures move through the United States district courts as quickly as possible.

Service of process

The delivery of the complaint and summons to a defendant by an authorized person.

Roundtable method

A method of setting reserves by using the consensus of two or more claims personnel who have independently evaluated the claims file.

Expert system method

A method of setting reserves with a software application that estimates losses and loss adjustment expenses.

Predictive model

A model used to predict an unknown outcome by means of a defined target variable.

General damages

A monetary award to compensate a victim for losses, such as pain and suffering, that do not involve specific measurable expenses.

Internet of Things (IoT)

A network of objects that transmit data to computers.

Internet of Things (IoT)

A network of objects that transmit data to each other and to central hubs through the internet.

Loss ratio

A ratio that measures losses and loss adjustment expenses against earned premiums and that reflects the percentage of premiums being consumed by losses

Direct loss

A reduction in the value of property that results directly and often immediately from damage to that property.

Staged accident

An accident deliberately caused by a person who intends to feign injury and collect on the ensuing claim.

Machine learning

Artificial intelligence in which computers continually teach themselves to make better decisions based on previous results and new data.

Reserve

The amount the insurer estimates and sets aside to pay on an existing claim.

Data mining

The analysis of large amounts of data to find new relationships and patterns that will assist in developing business solutions.

Complaint

The allegations made by a plaintiff in a lawsuit.

Average value method

A case reserving method that establishes a predetermined dollar amount of reserve for each claim as it is reported.

Allegation

A claim made in the complaint by the plaintiff, specifying what the plaintiff expects to prove to obtain a judgment against the defendant.

Complex claim

A claim that contains one or more characteristics that cause it to cost more than the average claim.

Coinsurance clause

A clause that requires the insured to carry insurance equal to at least a specified percentage of the insured's property value.

Contributory negligence

A common-law principle that prevents a person who has been harmed from recovering damages if that person's own negligence contributed in any way to the harm.

Comparative negligence

A common-law principle that requires both parties to a loss to share the financial burden of the bodily injury or property damage according to their respective degrees of fault.

Blockchain

A distributed digital ledger that facilitates secure transactions without the need for a third party.

Special Investigation Unit (SIU)

A division set up to investigate suspicious claims, premium fraud, or application fraud.

Summons

A document that directs a sheriff or another court-designated officer to notify the defendant named in the lawsuit that a lawsuit has been started and that the defendant has a specified amount of time to answer the complaint.

Material fact

A fact that is significant to a decision or matter at hand.

Misrepresentation

A false statement of a material fact on which a party relies.

Special damages

A form of compensatory damages that awards a sum of money for specific, identifiable expenses associated with the injured person's loss, such as medical expenses or lost wages.

Statute of limitations

A law that stipulates the length of time after an event during which legal proceedings (such as a lawsuit or criminal charges) may be initiated.

Indirect loss

A loss that arises as a result of damage to property, other than the direct loss to the property.

Information gain

A measure of the predictive power of one or more attributes.

Formula method

A method of setting claim reserves by using a mathematical formula.

Individual case method

A method of setting reserves based on the claim's circumstances and the claim representative's experience in handling similar claims.

Concealment

An intentional failure to disclose a material fact.

Punitive damages (exemplary damages)

A payment awarded by a court to punish a defendant for a reckless, malicious, or deceitful act to deter similar conduct; the award need not bear any relation to a party's actual damages.

Compensatory damages

A payment awarded by a court to reimburse a victim for actual harm.

Tortfeasor

A person or organization that has committed a tort.

Claims representative

A person responsible for investigating, evaluating, and settling claims.

Actuary

A person who uses mathematical methods to analyze insurance data for various purposes, such as to develop insurance rates or set claim reserves.

Discovery

A pretrial exchange of all relevant information between the plaintiff and defendant.

Motion for summary judgment

A pretrial request asking the court to enter a judgment when no material facts are in dispute and the moving party is entitled to judgment as a matter of law.

Motion in limine

A pretrial request that certain evidence be excluded from the trial.

Classification tree

A supervised learning technique that uses a structure similar to a tree to segment data according to known attributes to determine the value of a categorical target variable.

Leaf node

A terminal node of a classification tree that is used to classify an instance based on its attributes.

Tort

A wrongful act or omission, other than a crime or a breach of contract, that invades a legally protected right.

Hard fraud

Actions that are undertaken deliberately to defraud.

Cross-Industry Process for Data Mining (CRISP-DM)

An accepted standard for the steps in any data mining process used to provide business solutions.

Pro rata contribution

An approach to other insurance by which the insurers contribute to the loss payment in the proportion to which they contribute to the total amount of coverage purchased (their limits of liability).

Data science

An interdisciplinary field involving the design and use of techniques to process very large amounts of data from a variety of sources and to provide knowledge based on the data.

Insurable interest

An interest in the subject of an insurance policy that is not unduly remote and that would cause the interested party to suffer financial loss if an insured event occurred.

Algorithm

An operational sequence used to solve mathematical problems and to create computer programs.

Third Party Administrator (TPA)

An organization that provides administrative services associated with risk financing and insurance.

Public Adjuster

An outside organization or person hired by an insured to represent the insured in a claim in exchange for a fee.

Insurance fraud

Any deliberate deception committed against an insurer or an insurance producer for the purpose of unwarranted financial gain.

Producer

Any of several kinds of insurance personnel who place insurance and surety business with insurers and who represent either insurers, insureds, or both

Training data

Data that is used to train a predictive model and that therefore must have known values for the target variable of the model.

Soft fraud (opportunity fraud)

Fraud that occurs when a legitimate claim is exaggerated

Precision

In model performance evaluation, a model's correct positive predictions divided by its total positive predictions.

Accuracy

In model performance evaluation, a model's correct predictions divided by its total predictions.

Lift

In model performance evaluation, the percentage of positive predictions made by the model divided by the percentage of positive predictions that would be made in absence of the model.

Holdout data

In the model training process, existing data with a known target variable that is not used as part of the training data.

Domain knowledge

Information related to the context of the information a data scientist is working with.

Premium Audit

Methodical examination of a policyholder's operations, records, and books of account to determine the actual exposure units and premium for insurance coverages already provided

Big data

Sets of data that are too large to be gathered and analyzed by traditional methods.

Recursively

Successively applying a model.

Loss adjustment expense (LAE)

The expense that an insurer incurs to investigate, defend, and settle claims according to the terms specified in the insurance policy.

Root node

The first node in a classification tree.

First Notice of Loss (FNOL)

The initial report notifying the insurer of a claim.

Stare decisis

The principle that lower courts must follow precedents set by higher courts.

Ethics

The study of what constitutes good and bad behavior, dealing with moral duty and obligation.

Telematics

The use of technological devices to transmit data via wireless communication and GPS tracking.

Indemnify

To restore a party who has sustained a loss to the same financial position that party held before the loss occurred