Abstract: A Hellinger decision tree can detect fraudulent transactions in a data set of financial transactions. Applying the Hellinger decision tree uses a Hellinger distance. The Hellinger decision tree can be part of a machine learning algorithm. In an example, the Hellinger decision tree is a positive and unbalanced Hellinger decision tree used with an imbalanced positive and unlabeled data.

Abstract: The current invention pertains to a computer-implemented method, a method, a use thereof, a computer system and a computer program product for determining a potential for anomalous activity by a suspect user, particularly for detecting fraud and most particularly for detecting money laundering.

Abstract: The current invention relates to a computer-implemented method for determining a category of a request provided by a user by means of a user device; said user device comprising connection means and means for receiving a request description relating to said request from said user; the method comprising a plurality of steps among which: receiving, from said user, said request description, by means of said device; uploading said request description to a server, said server having access to a database comprising a plurality of previously categorized requests each comprising a category and a vocabulary comprising a plurality of word vector representations; identifying, by said server, a plurality of component words belonging to a natural language text string comprised in said request description; obtaining, for at least one of said plurality of component words, an associated word vector representation from said vocabulary; and determining a request vector, based on said at least one obtained word vector representa

Abstract: The current invention pertains to a computer-implemented method, a method, a use thereof, a computer system and a computer program product for determining a potential for anomalous activity by a suspect user, particularly for detecting fraud and most particularly for detecting money laundering.

Abstract: The current invention pertains to a computer-implemented method, a method, a use thereof, a computer system and a computer program product for determining a potential for anomalous activity by a suspect user, particularly for detecting fraud and most particularly for detecting money laundering.

Abstract: A database includes a plurality of physical entity records of insurance claims, and each physical entity record includes physical data values for at least one numeric attribute and partition-specifying values concerning values for one or more nominal attributes from one or more insurance claim records. A method for detecting anomalies in the database includes retrieving and partitioning the plurality of physical entity records from the database, training an unsupervised anomaly detection algorithm on the plurality of physical entity records to obtain a trained anomaly detection model for each partition, calculating an anomaly score for each physical entity record using the trained anomaly detection model associated with each partition, and updating the plurality of physical entity records in the database by adding the associated anomaly score. The method is used to determine if a user-provided physical entity record is fraudulent using the anomaly score.

Abstract: The current invention relates to a computer-implemented method for creating an overlay map, preferably a heat map, comprising: receiving at a server a user request for creation of the overlay map; loading vector data at the server comprising location data, the location data comprising at least one attribute; converting the vector data to image data composed of pixels according to an index scale; applying the converted image data to a color ramp; and creating the overlay map based on the converted image data and the color ramp; wherein said vector data is organized according to a plurality of layers, the location data comprising at least one attribute for each layer.

Abstract: The present disclosure relates to a method for detecting anomalies with respect to a database comprising a plurality of physical entity records of insurance claims, each physical entity record comprising physical data values for at least one numeric attribute and partition-specifying values concerning values for one or more nominal attributes from one or more insurance claim records. The method includes retrieving and partitioning the plurality of physical entity records from the database, training an unsupervised anomaly detection algorithm on the plurality of physical entity records to obtain a trained anomaly detection model for each partition, calculating an anomaly score for each physical entity record using the trained anomaly detection model associated with each partition, and updating the plurality of physical entity records in the database by adding the associated anomaly score. The method is used to determine if a user-provided physical entity record is fraudulent using the anomaly score.

Abstract: The current invention relates to a system for detecting anomalies, said system comprising a communication module having access to a database comprising a plurality of physical entity records, each physical entity record comprising physical data values for at least one numeric attribute and partition-specifying values concerning values for one or more nominal attributes; a computing device comprising a processor, tangible non-volatile memory, program code present on said memory for instructing said processor; wherein the communication module is arranged to provide said computing device access to said database, and wherein said computing device is configured for carrying out a method for calculating an anomaly score for each of said plurality of physical entity records.

Abstract: The current invention relates to a computer-implemented method for determining a category of a request provided by a user by means of a user device; said user device comprising connection means and means for receiving a request description relating to said request from said user; the method comprising a plurality of steps among which: receiving, from said user, said request description, by means of said device; uploading said request description to a server, said server having access to a database comprising a plurality of previously categorized requests each comprising a category and a vocabulary comprising a plurality of word vector representations; identifying, by said server, a plurality of component words belonging to a natural language text string comprised in said request description; obtaining, for at least one of said plurality of component words, an associated word vector representation from said vocabulary; and determining a request vector, based on said at least one obtained word vector representa

Abstract: The current invention pertains to a computer-implemented method, a method, a use thereof, a computer system and a computer program product for determining a potential for anomalous activity by a suspect user, particularly for detecting fraud and most particularly for detecting money laundering.

Abstract: The current invention relates to a computer-implemented method for creating an overlay map, preferably a heat map, comprising: receiving at a server a user request for creation of the overlay map; loading vector data at the server comprising location data, the location data comprising at least one attribute; converting the vector data to image data composed of pixels according to an index scale; applying the converted image data to a color ramp; and creating the overlay map based on the converted image data and the color ramp; wherein said vector data is organized according to a plurality of layers, the location data comprising at least one attribute for each layer.

Abstract: The current invention relates to a computing system for automated investment advice with at least one computer-readable medium comprising a database. The database comprises investment-related data such as instrument knowledge data relating to historical performance of a plurality of financial instruments and client knowledge data relating to client risk that includes a plurality of pre-defined risk categories. The computing system is configured to carry out a method for calculating a risk-adhering portfolio. The method includes calculating the risk-adhering portfolio comprising a selection of at least one of the plurality of financial instruments. The calculating relates to maximizing an expected value of a portfolio return and/or minimizing an expected risk while adhering to the assigned risk category by comparing a value relating to the risk-adhering portfolio to the at least one pre-defined risk parameter value.

Abstract: A computer-implemented method determines a category of a request provided by a user by means of a user device. The user device includes connection means and means for receiving a request description relating to said request from said user. The method includes receiving, from the user, the request description, by means of the device, and uploading the request description to a server. The server has access to a database which includes a number of previously categorized requests each including a category and a vocabulary, which includes a number of word vector representations. The method further includes identifying, by the server, a number of component words belonging to a natural language text string included in the request description; obtaining, for at least one of the component words, an associated word vector representation from the vocabulary, and determining a request vector, based on at least one obtained word vector representation.

Abstract: The current invention relates to a system for detecting anomalies, said system comprising a communication module having access to a database comprising a plurality of physical entity records, each physical entity record comprising physical data values for at least one numeric attribute and partition-specifying values concerning values for one or more nominal attributes; a computing device comprising a processor, tangible non-volatile memory, program code present on said memory for instructing said processor; wherein the communication module is arranged to provide said computing device access to said database, and wherein said computing device is configured for carrying out a method for calculating an anomaly score for each of said plurality of physical entity records.

Abstract: The current invention relates to a computing system for obtaining a privacy-filtered response to a query of a user, the computing system comprising a server, the server comprising a server processor, tangible non-volatile server memory, server program code present on said server memory for instructing said server processor; a computer-readable medium, the computer-readable medium comprising a database, said database comprising privacy settings comprising a privacy threshold; a device, said device comprising a device processor, tangible non-volatile device memory, device program code present on said device memory for instructing said device processor; wherein said server is configured for receiving raw transaction data from an external source such as a raw transaction database or a raw transaction feed.

Abstract: The current invention relates to a method of identifying relevant records in a database, preferably of determining potentially fraudulent records in said database. A graphical user interface is provided to a user to display multiple inputs, to receive inputs from the user and to display results. A first detection strategy targeted to detect existing records from the database is defined. The first detection strategy comprises multiple first inputs that comprise at least one threshold, at least one detection method, at least one weighting factor, and at least one parameter. The first inputs can be individually displayed on said graphical user interface for each of the at least one detection method. The first inputs can preferably be individually set for each of the at least one detection method. A second detection strategy targeted to detect said existing records from the database is defined. The second detection strategy comprises multiple second inputs.

Abstract: The current invention relates to a computing system for allocating a discount to a mortgage request, said computing system comprising a server, the server comprising a processor, tangible non-volatile memory, program code present on said memory for instructing said processor, optionally a network interface; at least one computer-readable medium, said computer-readable medium comprising a plurality of discount rules, at least one price-sensitivity model parameter, at least one customer-lifetime-value model parameter and optionally a database; said computing system configured for carrying out a method for said allocating of said discount.

Abstract: The current invention relates to a computer-implemented method for determining a risk associated with a physical entity based on physical data, said physical entity relating to real estate, said risk relating to a probability of damage of said physical entity with respect to a physical parameter comprised in said physical data, said method comprising the steps of: receiving, from a user, a geographical address of said physical entity; receiving, from the user, an entity category relating to said physical entity, said entity category being one of a row house, a semi-detached house, a detached house or an apartment; automatically determining said risk based on at least on said entity category; automatically determining a set of geographical coordinates of said physical entity based on said geographical address; automatically retrieving at least one value of said physical parameter belonging to said physical data based on said geographical coordinates from a database.

Abstract: A method for generating synthetic data records which include datasets that capture state-based transitions, according to which a state transition family is randomly selecting, according to the distribution of samples between the different clusters of users and the context variables are randomly sampled according to their distribution within the chosen cluster. The relevant Markov Chains models are selected according to the sampled context and the initial state of the sequence is randomly selected according to the distribution of states. A random walk process is initialized on the graph models and the random walk is performed process on each context separately, assuming context independency. The cause condition of the current transition is sampled for each state transition, based on the distributions on the selected edge.