Automated collaborative behavior analysis using temporal motifs

摘要

Described is a system for automated collaborative behavior analysis using temporal motifs. The system receives an input documents and change log files of a collaborative media, where the documents are continuously edited by multiple authors and where edits are recorded in the change log files, such as Wikipedia. A type of editing behavior by the authors of a given document is identified, and the edits made to the document are analyzed. The system reports how the authors interacted in a collaboration process, resulting in a set of reported author interactions. From the set of reported author interactions, a set of author interactions that are most and least significant in the collaboration process are identified. Then, based on the set of identified author interactions, future effects on documents of the collaborative media are estimated.

主权项

1. A system for automated collaborative behavior analysis, the system comprising:
one or more processors and a non-transitory memory having instructions encoded thereon such that when the instructions are executed, the one or more processors perform operations of:
receiving a plurality of documents and a plurality of change log files of a collaborative media as input, wherein the plurality of documents are continuously edited by a plurality of authors, and wherein edits are recorded in the plurality of change log files;identifying a type of editing behavior by authors of a given document in the plurality of documents;analyzing edits made to the given document;reporting how the authors interacted in a collaboration process, resulting in a set of reported author interactions;identifying, from the set of reported author interactions, a set of identified author interactions that are most and least significant in the collaboration process;estimating, based on the set of identified author interactions, future effects on documents of the collaborative media; andrepresenting the plurality of documents and the plurality of change log files as a temporal bipartite network, wherein the temporal bipartite network comprises a plurality of author type nodes, a plurality of page nodes, and a plurality of temporal edges annotating the time of the edits.