Computer-based systems having computer engines and data structures configured for machine learning data insight prediction and methods of use thereof

1 . A method, comprising:

receiving, by an inference engine, a query from a client terminal;

in response to receiving the query, executing instructions to cause the inference engine to perform operations comprising:

retrieving, from a database, a data set associated with the query;

generating, based on the data set, a decision tree model with a parent node and a plurality of child nodes, wherein the parent node is associated with the data set and the plurality of child nodes are associated with the parent node, and wherein the plurality of child nodes respectively correspond to a plurality of subsets of the data set, further comprising:

assigning a first insight comprising a sequence number associated with the data set into the parent node of the decision tree model, and

assigning, into a subset of the plurality of child nodes of the decision tree model, a second insight based on another sequence number that the subset of the plurality of child nodes follow from the parent node, wherein the second insight comprises the other sequence number and a rank number that are associated with a subset of the plurality of subsets of the data set, wherein the parent node and the plurality of child nodes have respectively corresponding sequence numbers, and wherein the subset of the plurality of child nodes have respectively corresponding rank numbers;

traversing the decision tree model based on the corresponding sequence numbers and the corresponding rank numbers to identify the parent node and the plurality of child nodes, wherein the traversing bypasses one or more nodes in the decision tree model that are not relevant to the query;

providing, to a natural language generator, the first insight assigned to the parent node and the second insight assigned to the subset of the plurality of child nodes to generate a textual summary of the data set;

in response to the textual summary being generated, compiling the textual summary of the data set, wherein the textual summary includes the first insight corresponding to the parent node, the second insight corresponding to the subset of the plurality of child nodes that have the respectively corresponding rank numbers, and a third insight corresponding to a comparison between the first insight and the second insight; and

generating a graphical user interface (GUI) including the textual summary of the data set for display at the client terminal.

2 . The method of claim 1 , wherein the sequence number indicates an order to traverse the parent node and the plurality of child nodes of the decision tree model.

3 . The method of claim 1 , wherein the rank number indicates a relevancy of the plurality of subsets of the data set associated with the parent node respectively corresponding to the plurality of child nodes to the query.

4 . The method of claim 1 , further comprising:

updating the GUI to include a graph representation of first transaction data of a first entity and second transaction data of a second entity, wherein the first transaction data is the data set associated with the parent node of the decision tree model.

5 . The method of claim 4 , wherein the graph representation includes a time series of transaction data related to the query.

6 . The method of claim 1 , wherein the query is a natural language text query related to spending behavior and wherein the data set includes transaction data.

7 . The method of claim 1 , further comprising:

determining a fourth insight corresponding to a course of action recommendation based on traversal of the decision tree model.

8 . The method of claim 1 , wherein, within the generated decision tree model, the parent node having a respectively corresponding sequence number is assigned with the first insight and the subset of the plurality of child nodes having the respectively corresponding rank numbers is assigned with the second insight, wherein the first insight is generated based on one or more historical insight requests, and wherein the second insight is generated based on the respectively corresponding sequence numbers.

9 . A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:

receiving, by the at least one computing device, a query from a client terminal;

in response to receiving the query, executing the instructions to cause the at least one computing device to perform the operations comprising:

retrieving, from a database, a data set associated with the query;

generating, based on the data set, a decision tree model with a parent node and a plurality of child nodes, wherein the parent node is associated with the data set and the plurality of child nodes are associated with the parent node, and wherein the plurality of child nodes respectively correspond to a plurality of subsets of the data set, further comprising:

assigning a first insight comprising a sequence number associated with the data set into the parent node of the decision tree model, and

assigning, into a subset of the plurality of child nodes of the decision tree model, a second insight based on another sequence number that the subset of the plurality of child nodes follow from the parent node, wherein the second insight comprises the other sequence number and a rank number that are associated with a subset of the plurality of subsets of the data set, wherein the parent node and the plurality of child nodes have respectively corresponding sequence numbers, and wherein the subset of the plurality of child nodes have respectively corresponding rank numbers;

traversing the decision tree model based on the corresponding sequence numbers and the corresponding rank numbers to identify the parent node and the plurality of child nodes, wherein the traversing bypasses one or more nodes in the decision tree model that are not relevant to the query;

providing, to a natural language generator, the first insight assigned to the parent node and the second insight assigned to the subset of the plurality of child nodes to generate a textual summary of the data set;

in response to the textual summary being generated, compiling the textual summary of the data set, wherein the textual summary includes the first insight corresponding to the parent node, the second insight corresponding to the subset of the plurality of child nodes that have the respectively corresponding rank numbers, and a third insight corresponding to a comparison between the first insight and the second insight; and

generating a graphical user interface (GUI) including the textual summary of the data set for display at the client terminal.

10 . The non-transitory computer-readable medium of claim 9 , wherein the sequence number indicates an order to traverse the parent node and the plurality of child nodes of the decision tree model.

11 . The non-transitory computer-readable medium of claim 9 , wherein the rank number indicates a relevancy of the plurality of subsets of the data set associated with the parent node respectively corresponding to the plurality of child nodes to the query.

12 . The non-transitory computer-readable medium of claim 9 , further comprising:

updating the GUI to include a graph representation of first transaction data of a first entity and second transaction data of a second entity, wherein the first transaction data is the data set associated with the parent node of the decision tree model.

13 . The non-transitory computer-readable medium of claim 9 , wherein the query is a natural language text query related to spending behavior and wherein the data set includes transaction data.

14 . The non-transitory computer-readable medium of claim 9 , wherein, within the generated decision tree model, the parent node having a respectively corresponding sequence number is assigned with the first insight and the subset of the plurality of child nodes having the respectively corresponding rank numbers is assigned with the second insight, wherein the first insight is generated based on one or more historical insight requests, and wherein the second insight is generated based on the respectively corresponding sequence numbers.

15 . A system, comprising:

a memory;

an inference engine;

at least one processor coupled to the memory and configured to perform operations comprising:

receiving, by the inference engine, a query from a client terminal;

in response to receiving the query, executing instructions to cause the inference engine to perform the operations comprising:

retrieving, from a database, a data set associated with the query;

generating, based on the data set, a decision tree model with a parent node and a plurality of child nodes, wherein the parent node is associated with the data set and the plurality of child nodes are associated with the parent node, and wherein the plurality of child nodes respectively correspond to a plurality of subsets of the data set, further comprising:

assigning a first insight comprising a sequence number associated with the data set into the parent node of the decision tree model, and

assigning, into a subset of the plurality of child nodes of the decision tree model, a second insight based on another sequence number that the subset of the plurality of child nodes follow from the parent node, wherein the second insight comprises the other sequence number and a rank number that are associated with a subset of the plurality of subsets of the data set, wherein the parent node and the plurality of child nodes have respectively corresponding sequence numbers, and wherein the subset of the plurality of child nodes have respectively corresponding rank numbers;

traversing the decision tree model based on the corresponding sequence numbers and the corresponding rank numbers to identify the parent node and the plurality of child nodes, wherein the traversing bypasses one or more nodes in the decision tree model that are not relevant to the query;

providing, to a natural language generator, the first insight assigned to the parent node and the second insight assigned to the subset of the plurality of child nodes to generate a textual summary of the data set;

in response to the textual summary being generated, compiling the textual summary of the data set, wherein the textual summary includes the first insight corresponding to the parent node, the second insight corresponding to the subset of the plurality of child nodes that have the respectively corresponding rank numbers, and a third insight corresponding to a comparison between the first insight and the second insight; and

generating a graphical user interface (GUI) including the textual summary of the data set for display at the client terminal.

16 . The system of claim 15 , wherein the sequence number indicates an order to traverse the parent node and the plurality of child nodes of the decision tree model.

17 . The system of claim 15 , wherein the rank number indicates a relevancy of the plurality of subsets of the data set associated with the parent node respectively corresponding to the plurality of child nodes to the query.

18 . The system of claim 15 , further configured to:

update the GUI to include a graph representation of first transaction data of a first entity and second transaction data of a second entity, wherein the first transaction data is the data set of the decision tree model.

19 . The system of claim 15 , wherein the query is a natural language text query related to spending behavior and wherein the data set includes transaction data.

20 . The system of claim 15 , wherein, within the generated decision tree model, the parent node having a respectively corresponding sequence number is assigned with the first insight and the subset of the plurality of child nodes having the respectively corresponding rank numbers is assigned with the second insight, wherein the first insight is generated based on one or more historical insight requests, and wherein the second insight is generated based on the respectively corresponding sequence numbers.