MACHINE LEARNING SYSTEMS FOR COLLABORATION PREDICTION AND METHODS FOR USING SAME
A machine learning system can include a data store and a computing device in communication with the data store. The data store can include entity data. The computing device can receive data describing at least one aspect of a position for the entity. The computing device can generate metadata for the position based on the data describing the at least one aspect, the metadata including a plurality of skills and tasks associated with the position. The computing device can identify task locations for the entity and determine a distribution of capacity across the same based on entity data. The computing device can determine physical proximity scores for each skill and task based on the metadata and the corresponding distribution of capacity. The computing device can generate a collaboration score for the position based on the plurality of physical proximity scores.
1 . A machine learning system, comprising:
a data store comprising entity data corresponding to an entity;
at least one computing device in communication with the data store, the at least one computing device being configured to:
receive data describing at least one aspect of a position for the entity;
generate metadata for the position based on the data describing the at least one aspect of the position, the metadata comprising a plurality of skills and tasks associated with the position;
identify a plurality of task locations for the entity;
determine a distribution of capacity across the plurality of task locations based on the entity data;
determine a plurality of physical proximity scores for each of the plurality of skills and tasks based on the metadata and the distribution of capacity across the plurality of task locations; and
generate a collaboration score for the position based on the plurality of physical proximity scores.
2 . The machine learning system of claim 1 , wherein the at least one computing device is further configured to generate the collaboration score via a trained machine learning model.
3 . The machine learning system of claim 2 , wherein the at least one computing device is further configured to generate the trained machine learning model by:
generating an initial machine learning model;
training, with a training dataset, the initial machine learning model to generate one or more experimental collaboration predictions, wherein the training dataset comprises historical entity data associated with the position and one or more known collaboration outcomes associated with the historical entity data;
determining an error of the initial machine learning model by comparing the one or more experimental collaboration predictions to the one or more known collaboration outcomes; and
generating a secondary machine learning model by adjusting the initial machine learning model based on the error, wherein the trained machine learning model is the secondary machine learning model.
4 . The machine learning system of claim 3 , wherein:
the initial machine learning model comprises a plurality of parameters and a first set of weight values that are applied to each of the plurality of parameters, wherein:
the plurality of parameters are based on the plurality of physical proximity scores; and
the first set of weight values determines a level of contribution of each of the plurality of parameters to the collaboration score; and
the at least one computing device is configured to generate the secondary machine learning model by:
determining at least one of the plurality of parameters that most contributed to the error;
adjusting one or more weight values of the first set of weight values that are associated with the at least one the plurality of parameters to generate a secondary set of weight values; and
generating the secondary machine learning model based on the plurality of parameters and the secondary set of weight values.
5 . The machine learning system of claim 1 , wherein the at least one computing device is configured to generate a report comprising the collaboration score.
6 . The machine learning system of claim 5 , wherein:
the at least one computing device is configured to determine at least one physical proximity score from the plurality of physical proximity scores that most positively contributed to the collaboration score; and
the report further comprises the at least one physical proximity score that most positively contributed to the collaboration score.
7 . The machine learning system of claim 5 , wherein:
the at least one computing device is configured to determine at least one physical proximity score from the plurality of physical proximity scores that most negatively contributed to the collaboration score; and
the report further comprises the at least one physical proximity score that most negatively contributed to the collaboration score.
8 . A machine learning method, comprising:
receiving, via at least one computing device, data describing at least one aspect of a position for an entity;
generating, via the at least one computing device, metadata for the position based on the data describing the at least one aspect of the position, the metadata comprising a plurality of skills and tasks associated with the position;
identifying, via the at least one computing device, a plurality of task locations for the entity;
determining, via the at least one computing device, a distribution of capacity across the plurality of task locations based on entity data corresponding to the entity;
determining, via the at least one computing device, a plurality of physical proximity scores for each of the plurality of skills and tasks based on the metadata and the distribution of capacity across the plurality of task locations; and
generating, via the at least one computing device, a collaboration score for the position based on the plurality of physical proximity scores.
9 . The machine learning method of claim 8 , further comprising generating the collaboration score by combining the plurality of physical proximity scores for each of the plurality of skills and tasks according to a predetermined weighting.
10 . The machine learning method of claim 8 , wherein the entity data comprising data describing a plurality of individuals associated with the entity, and the method further comprises anonymizing the entity data to remove identifying information corresponding to the plurality of individuals associated with the entity.
11 . The machine learning method of claim 8 , further comprising generating a collaboration classification for the position based on the collaboration score.
12 . The machine learning method of claim 11 , wherein the collaboration classification is generated according to:
c
(
x
i
j
g
)
=
{
position
least
likely
to
be
collaborative
if
h
(
x
i
j
g
)
<
h
0
position
may
be
collaborative
if
h
0
<
h
(
x
i
j
g
)
<
h
1
position
more
likely
to
be
collaborative
if
h
1
<
h
(
x
i
j
g
)
<
h
2
position
most
likely
to
be
collaborative
if
h
(
x
i
j
g
)
>
h
2
wherein:
h(x ijg ) is the collaboration score;
h 0 is a predefined collaborative position threshold;
h 1 is a predefined potentially collaborative position threshold;
h 2 is a predefined likely collaborative position threshold; and
c(x ijg ) is the collaboration classification.
13 . The machine learning method of claim 12 , further comprising:
transmitting the collaboration classification and the collaboration score to a collaboration application of a computing device; and
rendering, via the computing device, a user interface comprising the collaboration classification and the collaboration score.
14 . The machine learning method of claim 8 , further comprising collecting additional data describing the position from at least one external data source, wherein the metadata comprises the additional data describing the position.
15 . A non-transitory computer-readable medium embodying a program that, when executed by at least one computing device, causes the at least one computing device to:
receive data describing at least one aspect of a position for an entity;
generate metadata for the position based on the data describing the at least one aspect of the position, the metadata comprising a plurality of skills and tasks associated with the position;
identify a plurality of task locations for the entity;
retrieve, from a data store, entity data corresponding to the entity;
determine a distribution of capacity across the plurality of task locations based on entity data;
determine a plurality of physical proximity scores for each of the plurality of skills and tasks based on the metadata and the distribution of capacity across the plurality of task locations; and
generate a collaboration score for the position based on the plurality of physical proximity scores.
16 . The non-transitory computer-readable medium of claim 15 , wherein the program further causes the at least one computing device to generate the metadata by applying one or more natural language processing techniques that determine the plurality of skills and tasks associated with the position from the data describing the position.
17 . The non-transitory computer-readable medium of claim 15 , wherein the program further causes the at least one computing device to perform entity resolution on the data describing the at least one aspect of the position prior to generating the metadata.
18 . The non-transitory computer-readable medium of claim 15 , wherein the program further causes the at least one computing device to:
perform topic modeling on the data describing the at least one aspect of the position prior to generating the metadata, wherein the topic modeling generates one or more topics associated with the position; and
generate at least a portion of the metadata by retrieving historical entity data that matches the one or more topics associated with the position.
19 . The non-transitory computer-readable medium of claim 18 , wherein the historical entity data comprises historical skills and tasks associated with the position.
20 . The non-transitory computer-readable medium of claim 15 , wherein the program further causes the at least one computing device to generate the metadata by performing one or more keyword matching techniques to identify subsets of the data describing the at least one aspect of the position that describe the plurality of skills and tasks associated with the position.