Inferring meeting expense via a meeting expense and verification controller
This disclosure describes techniques that enable a Meeting Expense and Verification (MEV) controller to infer the meeting expense of a proposed meeting. The MEV controller may analyze a meeting request to identify meeting attendees, and in doing so, determine the meeting expense. In response to the meeting expense being greater than a predetermined meeting expense threshold, the MEV controller may suspend the meeting request.
1. A system, comprising:
one or more processors; and
memory coupled to the one or more processors, the memory including one or more modules that are executable by the one or more processors to perform operations to infer a meeting expense for a proposed meeting, calculate an actual meeting expense for an actual meeting, and recommend a change in meeting format, the operations to infer a meeting expense comprising:
intercepting, from a scheduler device, a meeting request for a proposed meeting of in-person meeting attendees at a first meeting location, wherein the first meeting location is a geolocation;
inferring a meeting expense for the proposed meeting based at least in part on the meeting request, wherein an inferring of the meeting expense includes:
determining time lost due to travel of the in-person meeting attendees to the first meeting location that contributes to lost work time (LWT) values of the in-person meeting attendees, the determining the time lost being performed at least using one or more trained machine-learning algorithms to infer a start location of each individual in-person meeting attendee, the machine-learning algorithms having been trained on historical calendar data of the each individual in-person meeting attendee and environmental factors that historically impact meeting at the first meeting location; and
in response to the inferred meeting expense being greater than a predetermined meeting expense threshold, suspending the meeting request, and
the operations to calculate an actual meeting expense for an actual meeting comprising:
for in-person meeting attendees attending the actual meeting at a meeting site:
interacting with sensors located at the meeting site to capture sensor data that includes video from the in-person meeting attendees during the meeting;
analyzing the sensor data, including comparing the sensor data to biometric profile data of the in-person meeting attendees, to identify in-person meeting attendees based on the comparison; and
determining an actual expense of the identified in-person meeting attendees based on their respective LWT values;
for virtual meeting attendees attending the meeting virtually via a virtual meeting platform:
interacting with the virtual meeting platform to capture attendance data that comprises Internet Protocol (IP) addresses associated with the virtual meeting attendees;
identifying the virtual meeting attendees from the IP addresses in the attendance data;
determining, based on historical profile data of the virtual meeting attendees identified in the attendance data, an expense of setting up the virtual meeting at a second meeting location for the virtual meeting attendees that contributes to LWT values of the virtual meeting attendees; and
determining an actual expense of the virtual meeting attendees based on their respective LWT values; and
calculating the actual meeting expense by aggregating the actual expense of the in-person meeting attendees and the actual expense of the virtual meeting attendees; and
the operations to recommend a change in meeting format comprising:
sensing virtual meeting connectivity of the virtual meeting attendees to the actual meeting;
receiving telemetry data that indicates Quality of Service (QOS) of the virtual meeting connectivity; and
recommending a change in meeting format from a virtual meeting to an in-person meeting for at least a subset of the virtual meeting attendees based on the telemetry data indicating that virtual meeting connectivity for the at least a subset of the virtual meeting attendees is below a predetermined QoS threshold.
2. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
retrieve, from an enterprise server, the biometric profile data associated with individual in-person meeting attendees.
3. The system of claim 1 , wherein the proposed meeting is scheduled to occur at a proposed meeting time, and wherein the one or more modules are further executable by the one or more processors to:
retrieve, from an enterprise server, an individual profile associated with a particular in-person meeting attendee of the proposed meeting;
determine an individual LWT value associated with the particular in-person meeting attendee at the proposed meeting time; and
in response to the individual LWT value being greater than the predetermined meeting expense threshold, determine an alternate meeting time to replace the proposed meeting time, based at least in part on the individual LWT value.
4. The system of claim 3 , wherein the one or more modules are further executable by the one or more processors to:
retrieve, from the enterprise server, historical device usage data associated with the particular in-person meeting attendee over a predetermined time interval;
analyze the historical device usage data to assign an effective worktime value to workday time increments; and
generate a data profile for the particular in-person meeting attendee, based at least in part on the worktime value assigned to the workday time increments,
wherein to determine the alternate meeting time is further based at least in part on the data profile.
5. The system of claim 4 , wherein the one or more modules are further executable by the one or more processors to:
determine the alternate meeting time to occur during a workday time increment with an assigned worktime value that is less than or equal to remaining workday time increments.
6. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
retrieve, from an enterprise server, a meeting attendee profile associated with a particular in-person meeting attendee of the proposed meeting;
determine an expense projection associated with the particular in-person meeting attendee at a proposed meeting time of the proposed meeting; and
based on the expense projection being greater than the predetermined meeting expense threshold, identify an alternate meeting attendee to attend the proposed meeting in place of the particular in-person meeting attendee.
7. The system of claim 6 , wherein the one or more modules are further executable by the one or more processors to:
analyze the meeting attendee profile to identify a plurality of supervised individuals that report to the particular in-person meeting attendee; and
determine an alternate expense projection for individual ones of the plurality of supervised individuals,
wherein the alternate meeting attendee corresponds to one of the plurality of supervised individuals, based at least in part on the alternate expense projection.
8. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
retrieve, from an enterprise server, a meeting attendee profile associated with a particular in-person meeting attendee of the proposed meeting;
parse the meeting request to identify a meeting subject-matter;
analyze the meeting attendee profile to identify a workgroup related to the meeting subject-matter; and
identify a plurality of members of the workgroup to attend the proposed meeting as alternate meeting attendees, based at least in part on the meeting subject-matter.
9. The system of claim 8 , wherein the one or more modules are further executable by the one or more processors to:
determine an alternate expense projection for individual ones of the plurality of members, and
wherein, the alternate meeting attendee corresponds to one of the plurality of members, based at least in part on the alternate expense projection.
10. The system of claim 1 , wherein the environmental factors include at least one of a vehicle parking availability at the first meeting location, walking distance from a vehicle parking location to the first meeting location, or public road traffic conditions en route to the first meeting location.
11. The system of claim 1 , wherein the one or more trained machine-learning algorithms include a supervised learning algorithm, an unsupervised learning algorithm, a semi-supervised learning algorithm, a naive Bayes algorithm, a Bayesian network algorithm, a decision tree algorithm, a neural network algorithm, a fuzzy logic model algorithm, or a probabilistic classification model algorithm.
12. A computer-implemented method performed under control of one or more processors, comprising:
inferring a meeting expense for a proposed meeting;
calculating an actual meeting expense for an actual meeting; and
recommending a change in meeting format,
the inferring of a meeting expense comprising:
intercepting, from a scheduler device, a meeting request for a proposed meeting of in-person meeting attendees at a first meeting location, wherein the first meeting location is a geolocation;
determining respective LWT values for the in-person meeting attendees, wherein the determining of respective LWT values includes determining time lost due to travel of the corresponding in-person meeting attendees to the first meeting location that contributes to the respective LWT value, the determining of the time lost being performed at least using one or more trained machine-learning algorithms to infer respective start locations of the in-person meeting attendees, the machine-learning algorithms having been trained on historical calendar data of the in-person meeting attendees and environmental factors that historically impact meeting at the first meeting location; and
in response to the inferred meeting expense being greater than a predetermined meeting expense threshold, suspending the meeting, and
the calculating of an actual meeting expense comprising, during a meeting of in-person attendees attending in person and virtual attendees attending virtually via a virtual meeting platform:
interacting with sensors located at the meeting site to capture sensor data that includes video from the in-person meeting attendees during the meeting;
analyzing the sensor data, including comparing the sensor data to biometric profile data of the in-person meeting attendees, to identify the in-person meeting attendees based on the comparison; and
determining an actual expense of the identified in-person meeting attendees based on their respective LWT values; and
for the virtual meeting attendees:
interacting with the virtual meeting platform to capture attendance data that comprises respective Internet Protocol (IP) addresses associated with the virtual meeting attendees;
identifying the virtual meeting attendees from the IP addresses in the attendance data; and
determining an expense of setting up the virtual meeting at a second meeting location for the virtual meeting attendees that contributes to LWT values of the virtual meeting attendees identified in the attendance data; and
determining an actual expense of the virtual meeting attendees based on their respective LWT values; and
calculating the actual meeting expense by aggregating the actual expense of the in-person meeting attendees and the actual expense of the virtual meeting attendees; and
the operations to recommend a change in meeting format comprising:
sensing virtual meeting connectivity of the virtual meeting attendees to the actual meeting;
receiving telemetry data that indicates Quality of Service (QOS) of the virtual meeting connectivity; and
recommending a change in meeting format from a virtual meeting to an in-person meeting for at least a subset of the virtual meeting attendees based on the telemetry data indicating that virtual meeting connectivity for the at least a subset of the virtual meeting attendees is below a predetermined QoS threshold.
13. The computer-implemented method of claim 12 , further comprising, for each of the in-person meeting attendees of the proposed meeting:
retrieving, from an enterprise server, historical calendar data associated with the in-person meeting attendee over a predetermined time interval; and
generating a data profile for the in-person meeting attendee based at least in part on the historical calendar data,
wherein determining the individual LWT values for the in-person meeting attendees of the proposed meeting is based at least in part on the data profile.
14. The computer-implemented method of claim 13 , further comprising:
determining an alternate meeting time for the proposed meeting, based at least in part on the individual LWT values for the in-person meeting attendees of the proposed meeting; and
transmitting, to the scheduler device, a notification of the alternate meeting time.
15. The computer-implemented method of claim 12 , further comprising:
determining that the individual LWT value of one of the in-person meeting attendees of the proposed meeting is greater than a predetermined worktime value threshold and, in response thereto:
identifying an alternate meeting attendee to attend the proposed meeting in place of the in-person meeting attendee.
16. The computer-implemented method of claim 15 , further comprising:
retrieving, from an enterprise server, a meeting attendee profile associated with the one of the in-person meeting attendees;
determining a meeting subject-matter, based at least in part on the meeting request; and
analyzing the meeting attendee profile to identify a workgroup associated with the meeting subject-matter,
wherein the alternate meeting attendee corresponds to a member of the workgroup.
17. The computer-implemented method of claim 12 , further comprising:
retrieving, from an enterprise server, a meeting attendee profile associated with one of the in-person meeting attendees of the proposed meeting; and
determining a travel time for the one of the in-person meeting attendees to travel to and from the first meeting location,
wherein determining the individual LWT value for the one of the in-person meeting attendees includes the travel time.
18. The computer-implemented method of claim 17 , further comprising:
determining that the individual LWT value for the one of the in-person meeting attendees is greater than a predetermined worktime value threshold; and
recommending, for the one of the in-person meeting attendees, a meeting format change from the in-person meeting to the virtual meeting, based on determining that the individual LWT value for the one of the in-person meeting attendees exceeds the predetermined worktime value threshold.
19. The computer-implemented method of claim 12 , wherein the environmental factors include at least one of a vehicle parking availability at the first meeting location, walking distance from a vehicle parking location to the first meeting location, or public road traffic conditions en route to the first meeting location.
20. The computer-implemented method of claim 12 , wherein the one or more trained machine-learning algorithms include a supervised learning algorithm, an unsupervised learning algorithm, a semi-supervised learning algorithm, a naive Bayes algorithm, a Bayesian network algorithm, a decision tree algorithm, a neural network algorithm, a fuzzy logic model algorithm, or a probabilistic classification model algorithm.