Computer-implemented method, device and system for generating and providing an interactive graphical user interface
An interactive graphical user interface is generated and provided. Transactional data records are received, each comprising a plurality of data attributes. Using the data attributes, each received data record is classified into zero, one or more classes of a first classification. Each previously classified data record is classified into zero, one or more classes of a second classification, using the data attributes and the first classification. A user selection of a class of the first classification is received and a text summary for the received data records having been attributed the selected class is generated. The text summary is fed to the graphical user interface. The text summary includes the class or classes attributed by the second classification attributed to the received data records having been attributed the selected class. A graphic representation for the selected class is generated.
1 . A computer-implemented method for generating, verifying and providing information in an interactive graphical user interface, comprising the steps of:
receiving, by at least one computing device, a plurality of transactional data records, each of the plurality of data records having a plurality of data attributes;
automatically classifying, by the at least one computing device:
at least one of the plurality of transactional data records into one or more classes of a first classification, as a function of the data attributes; and
the at least one of the plurality of transactional data records into one or more classes of a second classification, as a function of the data attributes and the first classification;
receiving, by the at least one computing device via the graphical user interface, a user selection of a class of the first classification;
dynamically generating, in response to the user selection, a plurality of text summaries, each including the one or more classes of the second classification for the at least one of the plurality of transactional data records having data attributes associated with the user selection, wherein generating each respective one of the text summaries includes:
automatically generating, by the at least one computing device, a prompt for a text summary of the selected class and the at least one class of the second classification;
providing the generated prompt to a text generator model; and
receiving, from the text generator model in response to the prompt, the generated respective text summary;
automatically verifying, by the at least one computing device, that data present in each of the plurality of generated text summaries are present in all of the plurality of generated text summaries;
providing, by the at least one computing device, at least one of plurality of the generated text summaries via the graphical user interface;
providing, via the graphical user interface, a graphic representation for the selected class;
providing, by the at least one computing device via the graphical user interface, the received plurality of transactional data records as tabular data; and
displaying, via the graphical user interface, the at least one of the plurality of the generated text summaries, the graphic representation, and the tabular data, including by highlighting part or parts of the at least one of the plurality of the generated text summaries, wherein the part or parts correspond to a class or classes from the second classification attributed to the received data records having been attributed the selected class.
2 . The method according to claim 1 , wherein generating the text summary further comprises generating a text summary of the class or classes attributed by the first classification and generating a text summary of the class or classes attributed by the second classification.
3 . The method according to claim 1 , further comprising displaying the classes obtained from the first classification.
4 . The method according to claim 1 , wherein the second classification is orthogonal in respect to the first classification.
5 . The method according to claim 1 , further comprising the display of classes obtained from the first classification, wherein first classification classes for which a preponderance of a class from the second classification are highlighted.
6 . The method according to claim 1 , further comprising highlighting part or parts of the generated graphic representation, in particular the part or parts being those corresponding to a class or classes from the second classification attributed to the received data records having been attributed the selected class, further in particular the part or parts being those corresponding to a class or classes from the second classification above a predetermined classification threshold.
7 . The method according to claim 1 , wherein the text generator model is an LLM, including a transformer-based model.
8 . The method according to claim 1 , wherein the first classification is based in a predetermined knowledge-area; and/or the second classification is based in a predetermined risk class.
9 . The method according to claim 1 , wherein classifying comprises applying a pretrained discriminative model, in particular the pretrained discriminative model comprising feature embedding pre-obtained from training transactional data records and corresponding classification.
10 . The method according to claim 1 , further comprising one or more of:
verifying if all numerical data present in the generated text summary is present in the received data records or present in numerical statistics computed from the received data records, in particular the numerical statistics comprising average, sum, median, deviation, minimum or maximum; and
previously including instructions in the prompt for explicitly generating precalculated numerical statistics computed from the received data records and verifying if all numerical data present in the generated text summary is present in the precalculated numerical statistics.
11 . A computer-implemented system for generating, verifying and providing information in an interactive graphical user interface, the system comprising:
at least one computing device including a memory accessing instructions stored on non-transitory processor readable media that, when executed configure the at least one computing device to:
receive a plurality of transactional data records, each of the plurality of data records having a plurality of data attributes;
automatically classify:
at least one of the plurality of transactional data records into one or more classes of a first classification, as a function of the data attributes; and
the at least one of the plurality of transactional data records into one or more classes of a second classification, as a function of the data attributes and the first classification;
receiving, by the at least one computing device via the graphical user interface, a user selection of a class of the first classification;
dynamically generating, in response to the user selection, a plurality of text summaries, each including the one or more classes of the second classification for the at least one of the plurality of transactional data records having data attributes associated with the user selection, wherein generating each respective one of the text summaries includes:
automatically generating, by the at least one computing device, a prompt for a text summary of the selected class and the at least one class of the second classification;
providing the generated prompt to a text generator model; and
receiving, from the text generator model in response to the prompt, the generated respective text summary;
automatically verifying, by the at least one computing device that data present in each of the plurality of generated text summaries are present in all of the plurality of generated text summaries;
providing, by the at least one computing device, at least one of plurality of the generated text summaries via the graphical user interface;
providing, via the graphical user interface, a graphic representation for the selected class;
providing, by the at least one computing device via the graphical user interface, the received plurality of transactional data records as tabular data; and
displaying, via the graphical user interface, the at least one of the plurality of the generated text summaries, the graphic representation, and the tabular data, including by highlighting part or parts of the at least one of the plurality of the generated text summaries, wherein the part or parts correspond to a class or classes from the second classification attributed to the received data records having been attributed the selected class.