Process flow diagram prediction utilizing a process flow diagram embedding

1. A method, comprising:

receiving a process flow diagram element of a process flow diagram;

identifying a context of the process flow diagram element, wherein the identifying a context comprises identifying (i) categories of elements connected to the process flow diagram element, (ii) swimlanes within the process flow diagram, and (iii) text included in the process flow diagram;

encoding features of the process flow diagram element into a semantic vector, wherein the features are identified from the context of the process flow diagram element and wherein the features comprise at least element type, swimlane, milestone, and text corresponding to the process flow diagram element; and

predicting, utilizing a process flow diagram embedding, a predicted process flow diagram element for the process flow diagram based upon the process flow diagram element, wherein the process flow diagram embedding receives and analyzes the features of the process flow diagram and outputs the predicted process flow diagram element, wherein the process flow diagram embedding ingests the semantic vector of the process flow diagram element, analyzes the semantic vector of the process flow diagram element against a training dataset, and outputs a semantic vector for the predicted process flow diagram element, wherein the predicting comprises converting the semantic vector for the predicted process flow diagram element into the predicted process flow diagram element, wherein the training dataset is generated by converting elements and contexts of the elements of previously developed process flow diagrams into training semantic vectors and aggregating the training semantic vectors into the training dataset.

2. The method of claim 1 , wherein the identifying element types of elements connected to the process flow diagram element comprises identifying the elements connected to the process flow diagram element, by analyzing a control flow of the process flow diagram.

3. The method of claim 1 , wherein the predicting comprises verifying an accuracy of the at least one process flow diagram element.

4. The method of claim 1 , comprising generating a process flow diagram catalog of previously developed process flow diagrams.

5. The method of claim 4 , comprising identifying a particular process flow diagram from the process flow diagram catalog, the particular process flow diagram being similar to the process flow diagram, wherein the identifying a particular process flow diagram from the process flow diagram catalog comprises (i) converting the previously generated process flow diagrams within the process flow diagram catalog into semantic vectors, (ii) comparing the semantic vectors of the process flow diagram against the semantic vectors of the previously developed process flow diagrams, and (iii) identifying a previously developed process flow diagram whose similarity to the process flow diagram exceeds a predetermined similarity threshold.

6. The method of claim 1 , wherein the identifying a context comprises representing process flow diagram elements within the process flow diagram as parts-of-speech, by utilizing linguistic processing.

7. The method of claim 1 , wherein the predicting comprises reconstructing a semantic vector for at least one of (i) a previous process flow diagram element and (ii) a subsequent process flow diagram element, in view of the semantic vector of the process flow diagram element.

8. The method of claim 7 , wherein the predicting comprises reverting the semantic vector of the predicted element into a process flow diagram element including (i) text, (ii) an identified swimlane, and (iii) an element type.

9. An apparatus, comprising:

at least one processor; and

a computer readable storage medium having computer readable program code embodied therewith and executable by the at least one processor, the computer readable program code comprising:

computer readable program code configured to receive a process flow diagram element of a process flow diagram;

computer readable program code configured to identify a context of the process flow diagram element, wherein the identifying a context comprises identifying (i) categories of elements connected to the process flow diagram element, (ii) swimlanes within the process flow diagram, and (iii) text included in the process flow diagram;

computer readable program code configured to encode features of the process flow diagram element into a semantic vector, wherein the features are identified from the context of the process flow diagram element and wherein the features comprise at least element type, swimlane, milestone, and text corresponding to the process flow diagram element; and

computer readable program code configured to predict, utilizing a process flow diagram embedding, a predicted process flow diagram element for the process flow diagram based upon the process flow diagram element, wherein the process flow diagram embedding receives and analyzes the features of the process flow diagram and outputs the predicted process flow diagram element, wherein the process flow diagram embedding ingests the semantic vector of the process flow diagram element, analyzes the semantic vector of the process flow diagram element against a training dataset, and outputs a semantic vector for the predicted process flow diagram element, wherein the predicting comprises converting the semantic vector for the predicted process flow diagram element into the predicted process flow diagram element, wherein the training dataset is generated by converting elements and contexts of the elements of previously developed process flow diagrams into training semantic vectors and aggregating the training semantic vectors into the training dataset.

10. A computer program product, comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code executable by a processor and comprising:

computer readable program code configured to receive a process flow diagram element of a process flow diagram;

computer readable program code configured to identify a context of the process flow diagram element, wherein the identifying a context comprises identifying (i) categories of elements connected to the process flow diagram element, (ii) swimlanes within the process flow diagram, and (iii) text included in the process flow diagram;

computer readable program code configured to encode features of the process flow diagram element into a semantic vector, wherein the features are identified from the context of the process flow diagram element and wherein the features comprise at least element type, swimlane, milestone, and text corresponding to the process flow diagram element; and

computer readable program code configured to predict, utilizing a process flow diagram embedding, a predicted process flow diagram element for the process flow diagram based upon the process flow diagram element, wherein the process flow diagram embedding receives and analyzes the features of the process flow diagram and outputs the predicted process flow diagram element, wherein the process flow diagram embedding ingests the semantic vector of the process flow diagram element, analyzes the semantic vector of the process flow diagram element against a training dataset, and outputs a semantic vector for the predicted process flow diagram element, wherein the predicting comprises converting the semantic vector for the predicted process flow diagram element into the predicted process flow diagram element, wherein the training dataset is generated by converting elements and contexts of the elements of previously developed process flow diagrams into training semantic vectors and aggregating the training semantic vectors into the training dataset.

11. The computer program product of claim 10 , wherein the identifying element types of elements connected to the process flow diagram element comprises identifying the elements connected to the process flow diagram element, by analyzing a control flow of the process flow diagram.

12. The computer program product of claim 10 , wherein the predicting comprises verifying an accuracy of the at least one process flow diagram element.

13. The computer program product of claim 10 , comprising generating a process flow diagram catalog of previously developed process flow diagrams.

14. The computer program product of claim 13 , comprising identifying a particular process flow diagram from the process flow diagram catalog, the particular process flow diagram being similar to the process flow diagram, wherein the identifying a particular process flow diagram from the process flow diagram catalog comprises (i) converting the previously generated process flow diagrams within the process flow diagram catalog into semantic vectors, (ii) comparing the semantic vectors of the process flow diagram against the semantic vectors of the previously developed process flow diagrams, and (iii) identifying a previously developed process flow diagram whose similarity to the process flow diagram exceeds a predetermined similarity threshold.

15. The computer program product of claim 10 , wherein the predicting comprises reconstructing a semantic vector for at least one of (i) a previous process flow diagram element and (ii) a subsequent process flow diagram element, in view of the semantic vector of the process flow diagram element.

16. The computer program product of claim 15 , wherein the predicting comprises reverting the semantic vector of the predicted element into a process flow diagram element including (i) text, (ii) an identified swimlane, and (iii) an element type.

17. A method, comprising:

training a machine-learning model, wherein the training comprises (i) receiving developed process flow diagrams, (ii) converting elements and contexts of the elements of the developed process flow diagrams into feature vectors, and (iii) generating a training dataset from the feature vectors, wherein the converting comprises identifying a context of process flow diagram elements within the process flow diagram and wherein the feature vectors are based upon the context of a given process flow diagram element;

receiving an undeveloped process flow diagram comprising at least one process flow diagram element;

converting the undeveloped process flow diagram into at least one feature vector, wherein the converting comprises (iv) identifying a context of the at least one process flow diagram element and (v) representing (a) the context and (b) the at least one process flow diagram element as a feature vector, wherein the feature vector comprises at least element type, swimlane, milestone, and text corresponding to the process flow diagram element; and

generating a prediction with respect to the undeveloped process flow diagram utilizing the at least one feature vector of the undeveloped process flow diagram, wherein the generating comprises (vi) providing the at least one feature vector to the machine-learning model and (vii) receiving the prediction from the machine-learning model, wherein the machine-learning model ingests the feature vector of the undeveloped process flow diagram, analyzes the feature vector of the undeveloped process flow diagram element against the training dataset, and outputs a predicted feature vector for the undeveloped process flow diagram, wherein the predicting comprises converting the predicted feature vector into a predicted process flow diagram element.