Proactive conflict resolution in node-based collaboration systems

1. A system comprising:

a processing hardware, a user interface (UI), and a system memory storing a software code and a conflict avoidance machine learning (ML) model trained using a deep reinforcement learning;

the processing hardware configured to execute the software code to:

receive from a first user, via the UI, project data describing a project including a plurality of tasks;

determine, using the project data, a node-based graph of the project including the plurality of tasks;

receive from one of the first user or a second user, via the UI, task action data producing a modification to the node-based graph;

identify, based on the modification to the node-based graph, a conflict among at least two nodes of the node-based graph;

predict, using the conflict avoidance ML model and the conflict, one or more candidate strategies for resolving the conflict;

identify, using the conflict avoidance ML model and based on the one or more candidate strategies in an automated process independent of the first user and the second user, a conflict avoidance strategy for resolving the conflict, wherein the conflict avoidance strategy comprises decoupling a node of the node-based graph;

display, via the UI, the node-based graph, a conflict alert indicating the node in the node-based graph and the conflict avoidance strategy to the one of the first user or the second user;

receive, via the UI, an input from the one of the first user or the second user to decouple the node indicated by the conflict alert;

in response to receiving the input, decouple the node indicated by the conflict alert;

display, via the UI, the node-based graph with the node decoupled from the node-based graph; and

further train the conflict avoidance ML model, using the deep reinforcement learning and the input from the one of the first user or the second user, to improve the conflict avoidance ML model in a feedback loop.

2. The system of claim 1 , wherein the system memory further stores a conflict database including a plurality of conflict resolution rules, and wherein the processing hardware is further configured to execute the software code to:

before identifying the conflict avoidance strategy, identify, based on a similarity of the conflict to a previously resolved conflict, one or more relevant rules for resolving the conflict from among the plurality of conflict resolution rules; and

wherein identifying the conflict avoidance strategy is further based on the one or more relevant rules.

3. A system comprising:

a processing hardware, a user interface (UI), and a system memory storing a software code and a conflict avoidance machine learning (ML) model trained using deep reinforcement learning;

the processing hardware configured to execute the software code to:

receive from a first user, via the UI, project data describing a project including a plurality of tasks;

determine, using the project data, a node-based graph of the project including the plurality of tasks;

receive from one of the first user or a second user, via the UI, task action data producing a modification to the node-based graph;

identify, based on the modification to the node-based graph, a potential future conflict among at least two nodes of the node-based graph;

predict, using the conflict avoidance ML model and the conflict, one or more candidate strategies for preventing the potential future conflict;

identify, using the conflict avoidance ML model and based on the one or more candidate strategies in an automated process independent of the first user and the second user, a conflict avoidance strategy for preventing the potential future conflict, wherein the conflict avoidance strategy comprises decoupling a node of the node-based graph;

display, via the UI, the node-based graph, a conflict alert indicating the node in the node-based graph and the conflict avoidance strategy to the one of the first user or the second user;

receive, via the UI, an input from the one of the first user or the second user to decouple the node indicated by the conflict alert;

in response to receiving the input, decouple the node indicated by the conflict alert;

display, via the UI, the node-based graph with the node decoupled from the node-based graph; and

further train the conflict avoidance ML model, using the deep reinforcement learning and the input from the one of the first user or the second user, to improve the conflict avoidance ML model in a feedback loop.

4. A method for use by a system including a processing hardware, a user interface (UI), and a system memory storing a software code and a conflict avoidance machine learning (ML) model trained using deep reinforcement learning, the method comprising:

receiving from a first user, by the software code executed by the processing hardware and via the UI, project data identifying a project including a plurality of tasks;

determining, by the software code executed by the processing hardware and using the project data, a node-based graph of the project including the plurality of tasks;

receiving from one of the first user or a second user, by the software code executed by the processing hardware and via the UI, task action data producing a modification to the node-based graph;

identifying, by the software code executed by the processing hardware and using the conflict avoidance ML model and based on the modification to the node-based graph, one of a conflict or a potential future conflict among at least two nodes of the node-based graph;

predicting, by the software code executed by the processing hardware and using the conflict avoidance ML model and the identified one of the conflict or the potential future conflict, one or more candidate strategies for one of resolving the conflict or preventing the potential future conflict;

identifying, by the software code executed by the processing hardware based on the one or more candidate strategies, in an automated process independent of the first user and the second user, a conflict avoidance strategy for the one of resolving the conflict or preventing the potential future conflict, wherein the conflict avoidance strategy comprises decoupling a node of the node-based graph;

displaying, by the software code executed by the processing hardware and via the UI, the node-based graph, a conflict alert indicating the node in the node-based graph and the conflict avoidance strategy to the one of the first user or the second user;

receiving, by the software code executed by the processing hardware and via the UI, an input from the one of the first user or the second user to decouple the node indicated by the conflict alert;

in response to receiving the input, decouple, by the software code executed by the processing hardware, the node indicated by the conflict alert;

in response to receiving the input, displaying, by the software code executed by the processing hardware and via the UI, the node-based graph with the node decoupled from the node-based graph; and

further training, by the software code executed by the processing hardware, the conflict avoidance ML model, using the deep reinforcement learning and the input from the one of the first user or the second user, to improve the conflict avoidance ML model in a feedback loop.

5. The method of claim 4 , wherein the system memory further stores a conflict database including a plurality of conflict resolution rules, the method further comprising:

identifying, by the software code executed by the processing hardware based on a similarity of the detected one of the conflict or the potential future conflict to a previously resolved conflict, one or more relevant rules for the one of resolving the conflict or preventing the potential future conflict from among the plurality of conflict resolution rules; and

wherein identifying the conflict avoidance strategy is further based on the one or more relevant rules.

6. The system of claim 1 , wherein the conflict avoidance strategy comprises reversing the modification to the node-based graph.

7. The system of claim 3 , wherein the conflict avoidance strategy comprises reversing the modification to the node-based graph.

8. The method of claim 4 , wherein the conflict avoidance strategy comprises reversing the modification to the node-based graph.