Machine learning (ML) system and method for optimizing meetings

Abstract

A machine-learning system and method for engineering and optimizing meetings. The method includes defining a meeting agenda comprising agenda items, generating prompts to solicit structured participant inputs, and enforcing contribution thresholds to ensure sufficient input. Participant inputs are integrated with internal organizational information and external information sources and analyzed using machine-learning techniques to generate results for each agenda item, including predictions, recommendations, and other optimized meeting products. Each completed meeting is stored as a structured dataset, and machine learning is applied across multiple completed meeting datasets over time to generate emergent knowledge and build an organization foundation model. The organization foundation model supports evaluation of participant and information source contributions, informs subsequent meetings, and enables time-based analysis. The system enables asynchronous, data-driven meetings that improve efficiency, accountability, and knowledge generation within and across organizations.

Description

Attorney Docket No.: 11214-011W01MACHINE LEARNING (ML) SYSTEM AND METHOD FOR OPTIMIZING MEETINGSCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 736,698, filed December 20, 2024, entitled " Machine Learning (ML) System and Method for Optimizing Meetings," the disclosure of which is expressly incorporated herein by reference in its entirety.BACKGROUND

[0002] The traditional meeting is a ubiquitous activity across all organizations. However, meetings are characterized frequently as being time consuming with questionable results. The traditional meeting also suffers from inefficient and questionable data capture, e.g., meeting minutes often inadequately capture the substance of a meeting; critical information is sometimes lost or overlooked in the limited capacity of a notetaker's attempt to "write down" the key ideas (which is subject to personal interpretation, e.g., a deliberation that occurs between the meeting chair and secretary). Recorded meetings offer little advantage, because all information is collected in raw form with no further analysis and interpretation.

[0003] Ingredients for an effective meeting include a thoughtful agenda, an effective chair, appropriate supporting documents and resources, and a clear goal; however, even these ingredients do not ensure a successful meeting as other dynamics can influence the engagement and voice of participants, i.e., input of the meeting participants.BRIEF DESCRIPTION OF THE DRAWINGS

[0004] The foregoing summary, as well as the following detailed description of illustrative implementations, is better understood when read in conjunction with the appended drawings. To illustrate the implementations, there are shown in the drawings example constructions; however, the implementations are not limited to the specific methods and instrumentalities disclosed. In the drawings:

[0005] FIG. 1 is a schematic diagram illustrating a system supporting operation of aspects of the disclosure;Attorney Docket No.: 11214-011W01

[0006] FIG. 2 is an operational flowchart illustrating an example method in accordance with the present disclosure; and

[0007] FIG. 3 is an example computing device in accordance with the present disclosure.DETAILED DESCRIPTION

[0008] The traditional meeting can be replaced with a machine-learning engineered meeting (ML-EM), a digital event created with a software application comprising tools and algorithms for measurably efficient and effective asynchronous meeting activity. The meeting is engineered (designed) in a software application around the elements of agenda, participants, pre-set input thresholds, purpose and anticipated products. Machine¬ learning is applied to participant input to generate results for each agenda item, identify and collect more information pertinent to those results, and conduct analysis and interpretation to yield optimized meeting products. Conclusion of the meeting can be achieved by vote or consensus of participants, or automatically with pre-set thresholds. An ML-EM is captured in its entirety as a dataset. The ML-EM coordinator can design subsequent meetings in response to apparent deficits from previous meetings; however, the ML-EM application itself can contribute to the design of subsequent meetings. The application is accessed by organizations with subscription rates tied to tiered product functionality, e.g., basic, professional, and premium (e.g., multi-unit linkage and B2B capability).

[0009] Glossary

[0010] As used herein the following terms have the following definitions:

[0011] A: An agenda; a set comprising all agenda items a for a meeting, where A = {a} = {a1, a2, a3...an}

[0012] o: An agenda item; an element of set A

[0013] B: A set comprising all input of a participant b for an agenda item a, where B= {b} = {b1, b2, b3...bn}

[0014] b: A participant's input for a prompt q; an element of set 8

[0015] c: An auxiliary (complementary) result derived intentionally from subset(s) of agenda items or from the agenda as a whole; an element of set YAAttorney Docket No.: 11214-011W01

[0016] CFM: Linking of informational assets between / among two or more organizations to achieve sufficient scale of data input to build a collaborative foundation model (CFM)

[0017] D: dark data; data with potential value that has been collected by an organization but has not been utilized to advance mission or goals

[0018] E: A set of external information assets, e.g., scientific or industry journals, market analyses / reports, public databases (e.g., sourced from the Department of Labor, CDC, NOAA, Census Bureau, etc.); also limited or full access to the foundational model of another (external) organization; E is a set comprising all external data and / or informational sources e; where E = {e} = {ei, e2, e3...en}

[0019] e: An external data and / or informational source; an element of set E

[0020] FM: foundation model

[0021] G: Organizational goals

[0022] / : A set of internal information assets, e.g., an organization's financial information, R& D data, KPIs, etc.; a set comprising all internal data and / or informational sources i, where / = { / } = { / i, it, ij—in}

[0023] i: An internal data and / or informational source; an element of set I

[0024] JOFM: A joint organization foundation model (JOFM), i.e., a larger scale machine-learning model trained on the combined OFMi-associated datasets of two or more organizations.

[0025] K: Emergent knowledge made possible by ML-EM, e.g., R, Y, KN, KO, and Kz

[0026] KN: Emergent knowledge yielded by machine learning applied to N.

[0027] Ko: Emergent knowledge yielded by machine learning applied to an organization foundation model (OFM) developed with ML-EM.

[0028] Kv: Machine learning applied to V is a source of emergent knowledge, Kv, that can incorporate time-series analysis.

[0029] Kz: Emergent knowledge yielded by machine learning applied to ZAttorney Docket No.: 11214-011W01

[0030] ML-EM: Machine-learning engineered meeting; a method for engineering (designing) meetings that enables the yield of emergent knowledge products at multiple levels within an organization and between / among organizations.

[0031] M: shorthand for ML-EM

[0032] M: A set of elements that represents a completed meeting, where M = {A, S, / , E, y}

[0033] N: A set of all completed meetings Mjfor an organization O over period T, where ∪ Mj= N

[0034] O: An organization

[0035] OFM: Organization foundation model developed with and supported by the ML-EM method.

[0036] P: the total input of a participant across A, i.e., all agenda items a, where ∪ Bi= P

[0037] Q: Q is a set comprising all agenda-item specific prompts q to solicit input,i.e., Q= {q} = {q1, q2,

[0038] q: a prompt q to solicit input for agenda item; an element of set Q

[0039] / ?: A set of results or meeting products for an agenda item generated by machine learning

[0040] r: A result for an agenda item o; an element for set R

[0041] S: The input of all participants for a meeting agenda, where u Pi= S

[0042] T: time

[0043] V: A set of all Niover multiple time periods, where ∪ Ni= V

[0044] W: Statistical measurement of the relative contribution / impact (weight) of an element (e.g., participant, unit, internal information asset) towards organizational success, e.g., measured as goals G or key performance indicators KPIs that underpin G.

[0045] Y: The total yield of a completed meeting M, i.e., the union of sets KA and YR equal Y, where YR U YA = YAttorney Docket No.: 11214-011W01

[0046] YA: A set of auxiliary (complementary) results c derived intentionally from subset(s) of agenda items or from the agenda as a whole, i.e., is a set comprising c, where YA= {c}= {c1, c2, c3...cn}.

[0047] YR: A set that represents the union of all R across A, i.e., the "yield" of all results across all agenda items, where YR= ∪ Ri

[0048] Detailed Description

[0049] With reference to FIG. 1, there is a schematic diagram illustrating a system 100 supporting operation of aspects of the disclosure that provides users of client devices 102a, 102b that execute a respective machine-learning engineered meeting (ML-EM) component 104a, 104b during the various levels in accordance with implementations of the present disclosure. As used herein, there may be three levels: an ML-EM level, an Organizational (O) level, and a business-to-business (B2B) level. Each level includes operations, as described below. Value derived from the ML-EM methods described herein is associated with significant new sources of emergent knowledge (e.g., generative Al products) that were not possible or obvious before, which can be understood by examining the detail of the ML-EM method as applied at the levels of meeting, organization, and business-to-business levels, i.e., the ML-EM (M) Level, the Organization (O) Level, and Business to business (B2B) Level, as described below.

[0050] The client device 102a may be a device within an end user site 101 and associated with, e.g., a customer. The client device 102a communicates over an internal LAN 103 with one or more end user internal data source(s) 105. The internal data sources 105 contain information about the end user's organization, such as financial information, key performance indicators (KPIs), research and development (R& D) information, etc. The client device 102a may communicate with a ML-EM server device 108 through the LAN 103 over a network 106, or the client device 102b may connect over an internal network directly with the ML-EM server device 108 (not shown). The ML-EM server device 108 may be physically or logically within a ML-EM site 107 associated with a ML-EM provider.

[0051] In system 100, a ML-EM server device 108, through the execution of a ML-EM internal component 110, performs operations that may include acquiring information from participants, internal data sources 112 and / or external sources 114. The internal data sources 112 may be similar to those described above; however, such data may beAttorney Docket No.: 11214-011W01associated with the ML-EM organization. The ML-EM server device 108 and internal component 110 may perform analysis of information to generate an organizational foundation model (OFM) 116 and / or joint organizational foundation model (JOFM) 118, as described below.

[0052] Users can interact with the system 120 using a dashboard user interface (not shown) running on the client device 102a, 102b. The client device 102a, 102b can represent a variety and / or combination of electronic devices capable of presenting the dashboard user interface and communicating with the ML-EM server device 108. The specific characteristics of the dashboard user interface and / or client device 102a, 102b can vary based upon the type of implementation of the system 120. For example, the dashboard user interface for a smartphone (as one or more of client device 102a, 102b) can vary in functionality and appearance from the dashboard user interface presented on a desktop computer (as one or more of client device 102a, 102b).

[0053] The system 100 can represent the hardware and / or software components necessary to provide the functionalities described herein with intelligent recommendations throughout all levels.

[0054] With reference to FIG. 2, there is an operational flow chart 200 illustrating an example method in accordance with the present disclosure.

[0055] Machine-Learning Engineered Meeting (ML-EM) Level

[0056] The ML-EM level begins at 202, when a meeting request is initiated. This may be initiated at the client device 102 using the client application component 104 and received at the server device 108. At 204, it is determined if the requester is authorized and if the request satisfies predetermined criteria, such as, but not limited to description and parameters for an organization (e.g., the customer associated with end user site 101). If it is determined that the meeting request satisfies authorization and meeting request description parameters, the flow advances to 206. If not, the request is denied and the operation flow returns to 202 to await a further meeting request.

[0057] At 206, the meeting is engineered using a combination of the client device 102, client application component 104, server device 108, and ML-EM internal component 110. The machine-learning engineered meeting (ML-EM, hereafter M) is the primary unit for soliciting participant input (P) and aggregating internal ( / ) and external informational (E)Attorney Docket No.: 11214-011W01assets that are subjected collectively to machine learning (e.g., deep learning) to yield optimized meeting products, e.g., results ( / ?). Each M includes an agenda A, where A is a set comprising all agenda items a:A = {a} = {ai, 02, 03... On}

[0058] At 208, it is determined if the engineered meeting meets quality assurance requirements. A coordinator tests to determine if all meeting elements have been met, and if so the operational flow continues at 210. If not, flow returns to 206 to redesign the meeting elements.

[0059] At 210, participant input is solicited. Above, each agenda item a is engineered (i.e., designed) to solicit input of a participant B. For each agenda item a there exists Q, where Q is a set comprising all prompts q to solicit input for agenda item aQ= {q} = {q1, q2, q3...qn}

[0060] For each agenda item a there exists B, where B is a set comprising all input of a participant b for agenda item atB= {b} = {bi, b2, b3...bn}

[0061] For each prompt q there is input of a participant b so that:n(Q) = n(B)

[0062] A participant can contribute an array of mixed (multi modal) inputs for each agenda item. Types of input include but are not limited to quantitative data (research results, statistical analyses, rankings, ratings), qualitative narrative in text form (opinion, ideas, descriptions, choices, suggestions, recommendations, preferences, reactions, decisions), identification of internal information assets (tagging I), and identification of external information assets (tagging E).

[0063] At 212, it is determined if the participant input contributions meet or exceed thresholds as defined for an organization. For example, it is determined if all specified sources of information have been submitted for the machine learning applications. A contribution threshold can be set for a participant for each agenda item a; contribution by participant is complete when this agenda-specific threshold is met or exceeded for all solicited input. The union of B across all agenda items a is U B and represents the total input of a participant across A, i.e., all agenda items atU B = PAttorney Docket No.: 11214-011W01

[0064] The union of P across all participants is U Ptand represents the total input of all participants 5 for agenda A:U Pi = S

[0065] If at 212, the thresholds are exceeded, the operational flow advances to 214. If not, then the operational flow returns to 210 to obtain additional inputs from the participants.

[0066] At 214, results are generated. The ML-EM application mines, aggregates, and integrates additional information drawing from internal (I) and external (E) information assets based on sources identified by participants and / or by sources identified by the specifications of M. Each M may include a set of internal data sources, where / is a set comprising all internal data and / or informational sources i, and may include a set of external data sources, where E is a set comprising all external data and / or informational sources e:I = {i} = {i1, i2, i3...in}E = {e} = {ei, e2, e3...en}

[0067] Machine learning across S, I, and E generates analyses, interpretation, and / or new content (e.g., for each a and across A = {a}) to yield optimized meeting products for each agenda item, i.e., a set of results R for each agenda item a, e.g., decisions, predictions, recommendations, new ideas. For each agenda item a there exists R, where R is a set comprising all results r:R = {r} = {r1, r2, r3...rn}

[0068] The union of R across agenda A is U Ri and represents the yield YR of a meeting; YR is the set of all R- for a meeting:

[0069] In addition to YR, auxiliary (complementary) results c may be derived intentionally (by design) with machine learning from subset(s) of agenda items or from the agenda as a whole, where YA is a set comprising all auxiliary results c:YA= {c} = {c1, c2, c3...cn}

[0070] The total yield V of a meeting is the union of the sets Y and YA'.W = YWhere Y and R are types of emergent knowledge (K) made possible by ML-EM.

[0071] Each completed M (M) is stored as a dataset, i.e., M is a set that includes A, S, / , E, and V:Attorney Docket No.: 11214-011W01M = {A, S, I, E, Y}

[0072] At 216, it is determined if ail meeting products, i.e. result that is intended by the design of the agenda have been created. If so they are distributed at 218. If not, the generation step at 214 is repeated such that missing or otherwise not-provided meeting products are generated.

[0073] At 220, it is determined if a meeting is completed through a consensus vote or other decision threshold. If it is determined that the meeting is completed than the operational flow continues to 222 where the completed meeting is submitted. Otherwise, the operational flow returns to 218.

[0074] Organization (O) Level

[0075] At 222, the Organization (O) Level activities begin where the union of M across all units in an organization is ∪ Mj= N and represents the entire set of completed meetings M for an organization for time period T; N is the set of all completed meetings Mjfor an organization O over period T:W

[0076] At 224, it is determined if quality assurance for completed meetings have met threshold standards. Otherwise, the operational flow returns to 222 to complete the meeting data sets. If yes at 224, then the completed meeting data set is stored at 226. Completed meetings M for an organization (O) can be organized in matrix form where each row (m) refers to the Ithunit within O and each column (n) refers to the jlhcompleted meeting M of that unit, arranged left to right in an ordinal sequence of columns. For example, indicates the fifth completed meeting (meeting number 5 or the fifth completed meeting) of the fourth unit (unit number 4 or the fourth unit). Although units might hold a different number of meetings over a time period, an organizational standard may implement monthly or quarterly meetings as a standard for all units, in which case all units would populate the same number of columns with M and at similar points of time.

[0077] Machine learning (e.g., deep learning) is applied to W to produce additional innovative information / knowledge (aka, emergent knowledge) KN such as (but not limited to) prediction, classification, evaluation, generative ideas, concepts, intellectual property. For each type of KN, the product is "time-stamped" and fully documented. KN can be employed to enhance organizational performance consistent with mission, goals, and vision.Attorney Docket No.: 11214-011W01

[0078] At 228, it is determined if there is sufficient accrual of completed meeting data sets for OFM development. If no at 228, the operational flow returns to 226. If yes, then the operational flow continues to 230 where the OFM is built. With increasing scale of cumulative data input, the ongoing application of machine learning (e.g., to A / ) can be utilized to build an Organization Foundation Model (OFM), developed with and supported by the ML-EM method. The OFM is a generative and innovative source of emergent knowledge Ko that is unique and proprietary (and thus a value asset) to an organization O; the yield of Ko can continue to dynamically evolve in at least two ways. One way is the ongoing application of machine learning to W can integrate additional or new sources (elements) for / and E and / or new ML algorithms. Another way is the ongoing application of machine learning may focus on A / as defined by a single time period T or may focus on multiple A / ; defined across multiple time periods T;; Vis the set of all A / ,- over multiple time periods Ti:∪ Ni= VMachine learning applied to V is a source of emergent knowledge, KV, that can incorporate time-series analysis.

[0079] Above, N, V, and an OFM can be utilized by an organization O to evaluate relative contributions W of elements (e.g., participants, units, information assets, meeting results, etc.) to progress towards organizational success, e.g., measured as goals G or key performance indicators KPIs that underpin G. The statistical impact or "weight" W of an element towards progress (a change in G or KPI) over time is measurable. For example, the relative contribution of input elements P, S, / , and E to outcomes Y, G, and KPI can be empirically evaluated. Furthermore, Y for each M can be evaluated for impact on organizational goals G and KPIs. W can be used to identify opportunities or resources for continuous improvement of the associated element (e.g., participants, units, information assets, meeting results, etc.). For example, the meeting input of a participant (employee) of an organization may be evaluated, providing motivation to improve measurable quality of participant input P for future M. Each participant has measurable value with reference to N, V, and OFM (and across the OFM Market; see below).

[0080] Over time, internal information assets / include, but are not limited to, Yield Y from previously completed meetings, M, a source of emergent knowledge that can be used in future meetings M. Thus, yield Y from completed meetings represents a form ofAttorney Docket No.: 11214-011W01internal information asset / within an organization that can be used across organizational units; and KN and Ko are also internal information assets / .

[0081] In the above, E may include the external information assets of other organizations (see also, B2B-level below). Over time, the application of ML-EM can be used to identify types of meetings within an organization with optimized recommended settings specific to meeting type. The engineering (designing) of M guides the mining in part of an organization's dark data D with intentionality and regularity, i.e., with innovative intentionality by the design of M and regularly with each occurrence M. This approach to mining D is a potential source of new information value Ko for an organization achieved with ML-EM.

[0082] Examples

[0083] Example 1: The Federal Bureau for Climate Research decides to implement machine-learning engineered meetings. FBCR is organized around multiple administrative units, specialized research teams, and various committees; each of these units conducts several meetings (M,) over a defined period of time T. ML-EM coordinators are identified to engineer (design) each meeting, paying attention to participants, agenda items, internal information assets, and external information assets. Each M is asynchronous, alleviating the burden of scheduling and significantly increasing the number of participants within FBCR who can contribute to the meeting. Each agenda item is specified using software to solicit a variety of input from each participant for each agenda item and includes the threshold for level of input by each participant for each agenda item. Participants might also be asked to "tag" sources of / and E, which can also be identified and added to the meeting by the ML- EM coordinator. For each agenda item, the coordinator also specifies the type(s) of results that are expected; once input from 5, 1, and E has been aggregated, machine learning is applied to generate a set of results R for each agenda item a across the agenda A, i.e., the yield of the meeting V. Each completed meeting M is stored as a dataset (a set) that includes 4, S, I, E, and V.

[0084] Example 2: The Global Business Corporation (GBC) conducts meetings over the course of one year (T=1 year) and has stored all datasets for completed M for this period, i.e., M;= N. Deep learning is applied to N to produce a substantially large Al model, aka, a foundation model (FM). This organization foundation model (OFM) is a source of new (emergent) knowledge (Ko) that was not possible before, which may include newAttorney Docket No.: 11214-011W01content, strategy, intellectual property, etc., that is the basis of new value for the organization, e.g., monetizing the licensing of new intellectual property or monetizing parameterized (tailored) access by other organizations to this OEM. The OEM is also used by HR to make decisions about recruitment, provide valuable performance feedback to employees, and custom design recommended professional development for each GBC employee; this saves money for GBC because it maximizes recruitment and retention. The GBC management team uses KN, KV, and Ko to intentionally implement plans of continuous improvement that is tailored to specific units based on the impact of their past contributions (W) over a time period T.

[0085] At 232, the interfaces between and among the OEMs are evaluated and negotiated. If acceptable, then at 234 Business to business (B2B) Level operations begin where the JOFM is built. An ML-EM application interface can link organization foundation models (OFM,) between / among two or more organizations, resulting in combinations of OEMs, where Z is a set comprising the union of two or more OFM,:o OFMi = Z

[0086] Kz represents the emergent knowledge produced by a joint organization foundation model (JOFM) resulting from combinations of OFMi, i.e., o OEM; = Z.

[0087] For example, given multiple OFMi across multiple organizations:* The OFMi of organization O, can be linked with OFM; of organization Oj to produce a new OFMij that results from deep learning underpinning a joint organization foundation model (JOFM), i.e., a larger scale machine-learning model trained on the combined datasets of both organizations.• The JOFM yields Kz that was not possible prior to the interfacing of OFM between organizations Oi and Oj.

[0088] OFMi developed using the ML-EM method across multiple organizations represents a network of OFM, i.e., the OFM Market, such that:• Based on potential combinations among OMF, across the OFM Market, different u OFMi = Z can be strategically developed (brokered) to build a variety of JOFM, e.g., to address larger-scale commercial interests, industry needs or initiatives.* The OFM Market represent a massive potential for emergent knowledge Kz and diverse applications based on large-scale combinations of OFM,.Attorney Docket No.: 11214-011W01

[0089] Application interfaces can also be created between organizations for any subordinately defined internal information assets / residing within these respective organizations, e.g., elements for meetings designed to optimize tracking systems for global vaccination distribution by a pharmaceutical company could be interfaced with similar meetings conducted by a major airline to optimize global tracking of luggage.

[0090] An ML-EM application interface can link informational assets between / among two or more organizations to achieve sufficient scale of data input to build a collaborative foundation model (CFM), e.g., informational assets among insurance companies, universities, private-public sector alliances.

[0091] Access of one business to another may be differential (i.e., limited or complete), balanced (O, and Oj both have the same level of complete or limited access to each other's OFM), or imbalanced (e.g., O, has complete access to Oj while Oj has limited access to O,).

[0092] Machine learning (e.g., deep learning) can be applied to any combination of OFMs developed between or among organizations, resulting in an emergent source of knowledge Kz derived from Z that would not have been possible otherwise. This knowledge has value.

[0093] Using set theory and a matrix framework as detailed here, access to elements of datasets between / among organizations can be defined precisely including range-defined time periods in which the elements were produced.

[0094] Examples

[0095] Example 3. Two global investment banks have implemented the ML-EM method. Each bank is heavily focused on financing for the energy sector while responding to major shifts in the geopolitical landscape, disruptions to global supply chains, and navigating long-term risk associated with uncertainties of climate change. Each organization utilizes the ML-EM approach to develop and maintain a proprietary organization foundation model. The two banks recognize the mutual value of partnering coilaboratively on financing sustainable energy infrastructure in Southeast Asia and strike a deal to build a joint organization foundation model (JOFM) to assist in developing, implementing, and evaluating long-term strategies in this market that reflects the strengths and expertise of both organizations. The JOFM adds value to the financial portfolios of both banks.Attorney Docket No.: 11214-011W01

[0096] Example 4. OFM, based on the ML-EM method are developed by several major public university systems across the country, who are each facing the challenge of responding to declining enrollments, changing demographics of students, and shifting priorities in workforce development. Each public university system is a multi-tiered organization comprising many functional "units." Recognizing a large-scale complex problem within regional variation, these public university systems decide to collaborate and build a joint organization foundation model with broad flexibility in applications to address these challenges. Based on the ML-EM method, this JOFM in part represents the strength (input) of thousands of administrators, staff, and faculty from across the country, all of whom have been involved in significant work focused on these issues over extended periods of time.

[0097] Sources of value with the L-EM method

[0098] There are several sources of potential financial value associated with ML-EM that were not possible before: Table 1. Ascending levels of ML-EM-derived emergent knowledge (K) associated with financial value (FV).Product Description of Source Description of Source Comments (value) Product regarding new sources of potential financial value (FV) Y The total yield of M Completed ML-engineered FV to an completed meeting, where organization meetings, where M = {A, S, / , E, / } derived from YR U YA = Y, the meeting results total yield of a (r). completedmeeting M.KN KN is a source of N N is the set of all M, for an FV to an emergent organization O over period organization knowledge of T: ^ Mi= N derived from different types emergent developed with knowledge of ML (e.g., deep different types learning) applied (KN). to N, e.g.,prediction,classification,evaluation,Attorney Docket No.: 11214-011W01generative ideas,concepts,intellectualpropertyKv Kv is a source of 1 / V is the set of all A / i for an FV to an emergent organization O over organization knowledge that multiple time periods Ti? derived from can incorporate Wi= V emergent time-series knowledge analysis. including timeseries analysis (Kv). Ko A generative and OFM Organization foundation • FV to an innovative source model (OFM) developed organization of emergent with ongoing ML applied to derived from knowledge Ko N or to V, such that the unique and that is unique and yield of Ko can continue to proprietary proprietary to an dynamically evolve in at organization organization O least two ways: foundation a. The ongoing model (Ko).application of machine • FV to an learning to N can organization, vis- integrate additional or a-vis access to new sources of / and E the OFM that and / or new ML can be algorithms monetized. b. The ongoing • FV to the ML-EM application of machine Corporation learning may focus onN or V.Kz Kz is a source of JOFM • Z is a set representing * FV to each Emergent the union of two or organization knowledge more OFMi (22) that • FV to the ML-EM derived from can be linked to build a Corporation combining two or JOFM.more OFM. * The OFM Marketrepresents thepotential network ofOFMi developed amongmultiple organizationsutilizing the ML-EMmethod.Table 1

[0099] Additional commercial value associated with the ML-EM method

[0100] ML-EM is implemented with a software application incorporating algorithms for machine learning to extract and generate emergent knowledge;Attorney Docket No.: 11214-011W01implementation will also require infrastructure, e.g., a cloud platform. The ML-EM method will create new job types, e.g., ML-EM consultant, and thus new career opportunities in the technology sector. The development, maintenance, ongoing evolution, and commercialization of this application will be under the authority of a corporation. ML-EM makes possible the ML-EM Corporation that will derive financial value in several ways that primarily fall under product, services, and transactions:item Description of commercial value Type Software Licensed use of product Product Consulting Training; optimizing implementation of product (technician, Service consultant)Brokering Negotiating and executing OFM interfacing across the OFM Service Market (broker)Access fee Tiered-scheduled access fee charged to Organization for OFM Transaction networkingStorage Archiving / V, V, OFM, CFM, and JOFM (cloud platform) Service Ko and Kz Proprietary emergent knowledge developed with individual Product OFM and across the OFM Market;The OFM The OFM Market can be configured as an investment market Service, Market focused on investors' perceived value of Ko and Kzas Product components of financial portfolios of participatingorganizations.Table 2

[0101] Establishing new industry standards using the ML-EM method

[0102] The ML-EM method is an example of human-centered Al that advances diversity, equity, and inclusion regarding input of participants in the meeting environment; no one person can monopolize a meeting and all individuals have fair and equal opportunity to contribute input in a non-threatening environment.

[0103] Methods of the present disclosure may provide empirical evidence of DEI such as, but not limited to, maximizing representation of human capital within an organization, mitigating organizational biases in evaluating solutions and strategies, and reducing liabilities (e.g., legal).

[0104] Methods of the present disclosure may measure relative contribution of individuals to team results, such as mitigating social loafing and optimizing professional development training to improve individual performance.

[0105] The ML-EM method establishes a new accountability for use of organizational resources, e.g., time and labor. For example, engineering a meeting andAttorney Docket No.: 11214-011W01getting measurable results, i.e., explicitly linking meeting input- output with results; significant reallocation of time and activity distribution for employees; irresponsible not to use ML-EM method, i.e., why would an organization not generate and capture this information, e.g., emergent knowledge Y, KN, KV, and KO? The information assets / and dark data D (value recovery) within an organization have potential financial value and the ML-EM method is an intentional application to fully utilize, mine, and repurpose these sources of information in an innovative way that was not possible before. The methods may provide identification of best sources of information over longer periods of time.

[0106] Emergent knowledge (e.g., generative Al products, IP) produced using the L-EM method will also be well protected, such as, preserving intellectual property at point of emergence. Activities will be fully documented, time stamped, and traceable to the relative contributions of the elements (e.g., participants, information sources, organizational units) that yielded the products.

[0107] The ML-EM method generates new sources of value in an organization that were previously unrecognized and inaccessible (see Tables 1 and 2).

[0108] It will likely become difficult over time for organizations to choose to dismiss the use of the ML- EM method, which has the potential to establish new industry standards for participation within an organization, full and systematic information utilization, protection of innovative knowledge products, and new sources of value for an organization's financial portfolio

[0109] FIG. 3 illustrates an example computer 300 that may include the kinds of software programs, data stores, and hardware that can implement event message processing, context determination, notification generation, and content delivery, as described above according to certain embodiments. As shown, the computing system 300 includes, without limitation, a central processing unit (CPU) 305, a network interface 315, a memory 320, and storage 330, each connected to a bus 317. The computing system 300 may also include an i / o device interface 310 connecting i / o devices 312 (e.g., keyboard, display and mouse devices) to the computing system 300. Further, the computing elements shown in computing system 300 may correspond to a physical computing system (e.g., a system in a data center) or may be a virtual computing instance executing within a computing cloud.Attorney Docket No.: 11214-011W01

[0110] The CPU 305 retrieves and executes programming instructions stored in the memory 320 as well as stored in the storage 330. The bus 317 is used to transmit programming instructions and application data between the CPU 305, I / O device interface 310, storage 330, network interface 315, and memory 320. Note, CPU 305 is included to be representative of a single CPU, multiple CPUs, a single CPU having multiple processing cores, and the like, and the memory 320 is generally included to be representative of a random access memory. The storage 330 may be a disk drive or flash storage device. Although shown as a single unit, the storage 330 may be a combination of fixed and / or removable storage devices, such as fixed disc drives, removable memory cards, optical storage, network attached storage (NAS), or a storage area-network (SAN).

[0111] Illustratively, the memory 320 includes one or more of the ML-EM client component 104 and ML-EM internal component 110, all of which are discussed in greater detail above. Further, storage 330 includes one or more of, agent item data (a) 331, prompt data (q) 332, participant data (p) 333, internal information source data (!) 110, external information source data (e) 112, auxiliary results data (c) 334, Organizational Foundation Model Data (OEM) 116, and Joint Organizational Foundation Model Data (OFM) 118, all of which are also discussed in greater detail above.

[0112] It should be understood that the various techniques described herein may be implemented in connection with hardware components or software components or, where appropriate, with a combination of both. Illustrative types of hardware components that can be used include field-programmable gate arrays (FPGAS), application-specific integrated circuits (ASICS), application-specific standard products (ASSPS), system-on-a-chip systems (SOCS), complex programmable logic devices (CPLDS), etc. The methods and apparatus of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as removeable drives (floppy diskettes, CD-ROMS), hard drives, including such on cloud-based environments, or any other machine-readable storage medium where, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the presently disclosed subject matter.

[0113] Although certain implementations may refer to utilizing aspects of the presently disclosed subject matter in the context of one or more stand-alone computer systems, the subject matter is not so limited but rather may be implemented in connectionAttorney Docket No.: 11214-011W01with any computing environment, such as a network or distributed computing environment. Still further, aspects of the presently disclosed subject matter may be implemented in or across a plurality of processing chips or devices, and storage may similarly be effected across a plurality of devices. Such devices might include personal computers, network servers, and handheld devices, for example.

[0114] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

[0115] Aspects of the present invention are described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to implementations of the disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0116] Although the figures show a specific order of method steps, the order of the steps may differ from what is depicted. Also, two or more steps may be performed concurrently or with partial concurrence. Such variation will depend on the software and hardware systems chosen and on designer choice. All such variations are within the scope of the disclosure. Likewise, software implementations could be accomplished with standard programming techniques with rule-based logic and other logic to accomplish the various connection steps, processing steps, comparison steps and decision steps.

[0117] It is to be understood that the methods and systems are not limited to specific synthetic methods, specific components, or to particular compositions. It is also to be understood that the terminology used herein is for the purpose of describing particular implementations only and is not intended to be limiting.Attorney Docket No.: 11214-011W01

[0118] As used in the specification and the appended claims, the singular forms "a," "an" and "the" include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from "about" one particular value, and / or to "about" another particular value. When such a range is expressed, another implementation includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent "about," it will be understood that the particular value forms another implementation. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint.

[0119] " Optional" or "optionally" means that the subsequently described event or circumstance may or may not occur, and that the description includes instances where said event or circumstance occurs and instances where it does not.

[0120] Throughout the description and claims of this specification, the word "comprise" and variations of the word, such as "comprising" and "comprises," means "including but not limited to," and is not intended to exclude, for example, other additives, components, integers or steps. " Exemplary" means "an example of" and is not intended to convey an indication of a preferred or ideal implementation. " Such as" is not used in a restrictive sense, but for explanatory purposes.

[0121] Disclosed are components that can be used to perform the disclosed methods and systems. These and other components are disclosed herein, and it is understood that when combinations, subsets, interactions, groups, etc. of these components are disclosed that while specific reference of each various individual and collective combinations and permutation of these may not be explicitly disclosed, each is specifically contemplated and described herein, for all methods and systems. This applies to all aspects of this application including, but not limited to, steps in disclosed methods. Thus, if there are a variety of additional steps that can be performed it is understood that each of these additional steps can be performed with any specific implementation or combination of implementations of the disclosed methods.

Claims

Attorney Docket No.: 11214-011W01WHAT IS CLAIMED IS:

1. A method for engineering meetings, comprising:determining agenda items for a meeting;receiving inputs from participants of the meeting in accordance with the agenda items;automatically integrating internal information related to an organization and external information outside the organization;analyzing the agenda items, the inputs, the internal information and the external information to generate results as predictions and recommendations for each agenda item;storing the results for a completed meeting;repeating the above for a plurality of completed meetings associated with the organization over a predetermined period of time; andlearning from completed meetings over the predetermined period of time to build an organization foundation model (OFM) that is a source of knowledge within the organization.

2. The method of claim 1, further comprising generating prompts that provided to participants to solicit the inputs.

3. The method of claim 2, further comprising setting an input threshold for each participant.

4. The method of claim 1, wherein the predictions and recommendations yield optimized meeting products for each agenda item.

5. The method of claim 1, further comprising storing the completed meeting as a dataset that includes A, S, I, E, and Y as M_ = {A, S, I, E, Y},wherein A is the agenda containing agenda items,wherein S is the input of the participants,wherein I is the internal information,wherein E is the external information, andV is a total yield of the completed meeting.Attorney Docket No.: 11214-011W016. The method of claim 1, further comprising using the OFM to evaluate contributions of the participants, the internal information, the external information and the results as key performance indicators (KPIs).

7. The method of claim 1, further comprising using the results as internal information in subsequent meetings.

8. The method of claim 1, further comprising linking OFM with an OFM for a second organization to generate a joint organization foundation model (JOFM).

9. The method of claim 8, further comprising developing the OFM from two or more JOFM.

10. The method of claim 1, further comprising measuring the predictions and recommendations.

11. A system for engineering meetings, comprising:at least one client device including a processor and memory storing a machinelearning engineered meeting (ML-EM) client component; anda server device comprising:a processor;a network interface configured to communicate with the at least one client device; anda memory storing an ML-EM internal component that, when executed by the processor, causes the server device to:engineer a meeting defined by an agenda comprising a plurality of agenda items; receive participant input for each agenda item from the at least one client device;integrate the participant input with internal organizational information and external information sources;Attorney Docket No.: 11214-011W01apply machine learning to the participant input, the internal organizational information, and the external information sources to generate results for each agenda item; andstore the generated results in a storage device as a completed meeting dataset.

12. The system of claim 11, wherein the ML-EM internal component is further configured to generate agenda-item-specific prompts and transmit the prompts to the at least one client device for solicitation of the participant input.

13. The system of claim 11, wherein the ML-EM internal component is further configured to enforce contribution thresholds defining a minimum level of participant input required for each agenda item prior to applying the machine learning.

14. The system of claim 11, wherein the completed meeting dataset is stored as a structured data object comprising:agenda data representing the agenda items;participant input data;internal information source data;external information source data; andmeeting yield data comprising results generated by the machine learning.

15. The system of claim 11, wherein the ML-EM internal component is further configured to aggregate a plurality of completed meeting datasets associated with an organization over a predetermined time period and apply machine learning to the aggregated datasets to generate emergent knowledge.

16. The system of claim 15, wherein the emergent knowledge is embodied in an organization foundation model (OFM) stored in the storage device and trained using the aggregated completed meeting datasets.Attorney Docket No.: 11214-011W0117. The system of claim 16, wherein the ML-EM internal component is further configured to use the organization foundation model to evaluate relative contributions of one or more of:participants,organizational units, orinformation sources,with respect to organizational goals or key performance indicators.

18. The system of claim 16, wherein the ML-EM internal component is further configured to interface the organization foundation model with a foundation model associated with a second organization to generate a joint organization foundation model (JOFM).

19. The system of claim 11, wherein the ML-EM internal component is further configured to designate results from the completed meeting dataset as internal organizational information for use in engineering a subsequent meeting.

20. The system of claim 11, wherein the server device further comprises a storage device configured to store one or more of:agenda item data,prompt data,participant input data,internal information source data,external information source data,auxiliary results data,organization foundation model data, andjoint organization foundation model data.

Images