Adaptive analytics user interfaces
AssigneeQliktech International AB
InventorsMatthew Kelly, Vesna Brown, Elif Tutuk
Life of this patent
Estimating the term…
Abstract
Provided are methods, systems, and apparatuses for enabling an analytics user interface to be adaptive based on, among other things, content of a user-defined query. A computing device may receive the query, which is to be applied to a dataset. The computing device way receive the query via an analytics user interface. Based on the query and using a recommendation engine, a plurality of recommended result elements and associated visualization elements may be determined. Based on the query and a selected recommended result element, a query result may be generated. The query result may contain a portion of records from the dataset, which may associated with the query and the at least one result element.
Claims18
A method comprising:
- receiving, by a computing device via an analytics user interface, a query to be applied to a dataset, wherein the query comprises at least one keyword;
- determining, based on the query and a machine learning module of a recommendation engine, a plurality of recommended result elements and associated visualization elements, wherein the plurality of recommended result elements comprises a plurality of filters to be applied to the dataset, and wherein the machine learning module is trained based on previous queries that comprise the at least one keyword and are associated with the plurality of filters;
- providing, via the analytics user interface, the plurality of recommended result elements and associated visualization elements;
- receiving, via the analytics user interface, a selection of at least one recommended result element of the plurality of recommended result elements, wherein the at least one recommended result element comprises at least one filter of the plurality of filters;
- determining, based on:
- the query, the at least one recommended result element, and a selection of at least one recommended visualization element, a query result, wherein the query result is associated with:
- the query, the at least one recommended result element, and the at least one recommended visualization element, and wherein the query result comprises a portion of the dataset resulting from the at least one filter applied to the dataset;
- and
- generating, based on the at least one recommended visualization element, a visualization of the query result.
A method comprising:
- (a) receiving, by a computing device via an analytics user interface, a query, comprising at least one keyword, and a result element, wherein the result element is one of a plurality of result elements and comprises at least one filter to be applied to a dataset;
- (b) determining, based on the query and the result element, a query result comprising a portion of the dataset resulting from application of the at least one filter on the dataset;
- (c) receiving a selection of a visualization element from a plurality of visualization elements to apply to the query result, wherein the plurality of visualization elements are provided at the analytics user interface based on the result element selected at the analytics user interface;
- (d) generating a training dataset based on:
- the at least one keyword of the query, the at least one filter associated with the result element, and the visualization element;
- and
- (e) training a machine learning module of a recommendation engine using the training dataset, wherein the machine learning module is trained to provide the at least one filter as a recommended result element for a further query that comprises the at least one keyword.
A non-transitory computer readable medium storing processor executable instructions that, when executed by at least one processor, cause the at least one processor to:
- receive, via an analytics user interface, a query to be applied to a dataset, wherein the query comprises at least one keyword;
- determine, based on the query and a machine learning module of a recommendation engine, a plurality of recommended result elements and associated visualization elements, wherein the plurality of recommended result elements comprises a plurality of filters to be applied to the dataset, and wherein the machine learning module is trained based on previous queries that comprise the at least one keyword and are associated with the plurality of filters;
- provide, via the analytics user interface, the plurality of recommended result elements and associated visualization elements;
- receive, via the analytics user interface, a selection of at least one recommended result element of the plurality of recommended result elements, wherein the at least one recommended result element comprises at least one filter of the plurality of filters;
- determine, based on:
- the query, the at least one recommended result element, and a selection of at least one recommended visualization element, a query result, wherein the query result is associated with:
- the query, the at least one recommended result element, and the at least one recommended visualization element, and wherein the query result comprises a portion of the dataset resulting from the at least one filter applied to the dataset;
- and
- generate, based on the at least one recommended visualization element, a visualization of the query result.
Dependencies are inferred from the claim text.