Multi-party conference scheduling method and system based on big data optimization

Abstract

The invention discloses a multi-party conference scheduling method and system based on big data optimization, and relates to the technical field of conference scheduling modeling and reasoning, and the method comprises the steps: collecting multi-source data, carrying out the unified time standardization and spatial geocoding processing of the multi-source data, constructing a scheduling state vector model, and carrying out the multi-party conference scheduling. A conference scheduling intention is deeply analyzed based on a natural language processing technology, a conference theme is compared with semantic features in historical conference data through a semantic matching mechanism, a multi-target scoring mechanism is introduced, a scheduling time period is evaluated, and a conference scheduling candidate scheme is pushed to a user for confirmation in a structured abstract form. According to the method, by constructing the multi-dimensional scheduling state vector and fusing the time availability, the space accessibility, the resource compatibility and the time preference, more accurate conference scheduling modeling and evaluation are achieved, the effectiveness and the actual performability of candidate time period screening are remarkably improved, and the problems that in a traditional method, the scheduling result is unreasonable, and the user acceptability is low are solved.

Description

Technical Field

[0001] The present invention relates to the field of conference scheduling modeling and reasoning technology, and specifically to a multi-party conference scheduling method and system based on big data optimization. Background Art

[0002] With the expansion of organizational collaboration and the prevalence of remote work, the demand for multi-party conference scheduling in cross-departmental, cross-regional, and cross-time zone scenarios is growing. Traditional conference scheduling methods mainly rely on calendar-based time matching, manual coordination, or fixed rule-based matching systems, which are difficult to cope with the complex challenges of resource coordination and personnel deployment. In recent years, with the development of big data processing, natural language processing, and multi-objective optimization algorithms, intelligent scheduling technology has been gradually applied to conference management systems, and has initially acquired the ability to automatically parse meeting requests, match participants' free time, and push candidate time slots. However, existing scheduling systems are still mainly in the rule-driven or static optimization stage, lacking in-depth semantic understanding and multi-dimensional constraint collaborative modeling, and are difficult to meet the comprehensive requirements of scheduling efficiency, adaptability, and user feedback responsiveness in real-world conference environments.

[0003] Existing multi-party conference scheduling technologies suffer from three core flaws. First, when modeling participant schedulability, most methods rely solely on Boolean temporal availability checks, ignoring key influencing factors such as spatial accessibility, resource compatibility, and behavioral preferences. These methods lack the ability to uniformly model multi-source heterogeneous data, making it difficult to accurately describe users' true scheduling capabilities. Second, when parsing meeting requirements, traditional methods primarily rely on structured field extraction, lacking a deep understanding of the semantic information in natural language meeting requests. They are unable to infer resource requirements and temporal patterns based on meeting agendas, resulting in scheduling recommendations that lack contextual sensitivity and behavioral orientation. Third, when optimizing scheduling, most existing systems employ static rules or single-objective algorithms for matching, failing to simultaneously optimize multiple objectives such as temporal overlap, resource utilization efficiency, and spatial cost-to-temporal preference alignment. Furthermore, when users report conflicts, overall scheduling often requires rescheduling, lacking a refined mechanism for handling local conflicts, resulting in inefficient responses and a poor user experience. In summary, current technical means are difficult to realize the integrated intelligent scheduling process of "scheduling state modeling - semantic perception reasoning - feedback adaptive optimization", and cannot achieve the comprehensive technical effects that can be achieved by the present invention in data fusion, demand expression and interactive optimization. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: the existing multi-party conference scheduling methods lack multi-dimensional modeling of scheduling status, insufficient ability to understand the semantics of conference requests, lack of feedback adaptability mechanism for scheduling optimization, and how to achieve multi-objective conference arrangements that take into account both scheduling efficiency and adaptability under the drive of multi-source data and user semantics.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a multi-party conference scheduling method based on big data optimization, comprising collecting multi-source data, performing unified time normalization and spatial geocoding processing on the multi-source data, and constructing a scheduling state vector model, which comprehensively considers four dimensions: time availability, spatial accessibility, resource compatibility, and time preference;

[0007] Using natural language processing technology, the system deeply analyzes meeting scheduling intentions, extracting the meeting topic, attendee list, meeting duration, resource requirements, and time expectations from free text. Using a semantic matching mechanism, the system compares the meeting topic with semantic features in historical meeting data, automatically inferring the meeting's resource allocation and common time preferences. Combined with the scheduling state vector model, the system outputs a set of candidate meeting scheduling solutions.

[0008] A multi-objective scoring mechanism is introduced to evaluate the scheduling time period, and the candidate meeting scheduling plans are pushed to the user for confirmation in the form of a structured summary. If there is a conflict in user feedback, the conflicting part is adjusted based on the minimum disturbance optimization mechanism.

[0009] As a preferred solution of the multi-party conference scheduling method based on big data optimization described in the present invention, the multi-source data includes the user's historical meeting records, personal schedule, office location and commuting range, conference equipment usage habits and time preference information.

[0010] As a preferred solution of the multi-party conference scheduling method based on big data optimization described in the present invention, the semantic matching mechanism includes using word embedding conversion technology to vectorize the conference topic keywords during the semantic parsing process, and combining the semantic distribution characteristics of historical meetings to perform similarity analysis between the current scheduling request and the existing conference types, and infer the resource requirements and time patterns of the meeting.

[0011] As a preferred solution of the multi-party conference scheduling method based on big data optimization described in the present invention, the output conference scheduling candidate plan set includes setting a conflict detection mechanism in the scheduling candidate plan generation process, including time conflict, space conflict and resource conflict, introducing a flexible scheduling boundary strategy, and performing tolerance judgment and dynamic intervention.

[0012] As a preferred solution of the multi-party conference scheduling method based on big data optimization described in the present invention, the evaluation of the scheduling time period includes dynamically weighting the priorities of the candidate time periods in combination with the historical behavior data of the users.

[0013] As a preferred solution of the multi-party conference scheduling method based on big data optimization described in the present invention, the multi-objective scoring mechanism includes time period overlap rate, personal time preference, spatial accessibility, conference resource availability and schedule fragmentation degree evaluation indicators, and all candidate solutions are weightedly sorted and prioritized.

[0014] As a preferred solution of the multi-party conference scheduling method based on big data optimization described in the present invention, the method of pushing the candidate conference scheduling plan to the user for confirmation in the form of a structured summary includes adopting a minimum disturbance optimization mechanism to adjust the time period or resource configuration when the user provides conflicting feedback on the candidate scheduling plan, and retaining the confirmed participant arrangements and resource allocation.

[0015] Another object of the present invention is to provide a multi-party conference scheduling system based on big data optimization, which can compare the conference theme with the semantic features in historical conference data through a semantic matching mechanism, thereby solving the problem that the current multi-party conference scheduling method has insufficient ability to understand the semantics of meeting requests.

[0016] As a preferred solution of the multi-party conference scheduling system based on big data optimization described in the present invention, it includes: a scheduling state modeling and demand semantic analysis module, a cross-available interval construction and multi-objective scheduling optimization module, and a scheduling result sorting feedback and local callback correction module; the scheduling state modeling and demand semantic analysis module is used to extract the input variables required for scheduling and establish a behavioral capability portrait of the participants; the cross-available interval construction and multi-objective scheduling optimization module is used to find the optimal conference arrangement time period and location configuration that meets the participants, resource constraints, and preference goals; the scheduling result sorting feedback and local callback correction module is used to sort according to the comprehensive satisfaction index and recommend it to the meeting initiator for selection through a guided interactive interface.

[0017] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a multi-party conference scheduling method based on big data optimization.

[0018] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a multi-party conference scheduling method based on big data optimization.

[0019] Beneficial effects of the present invention: The multi-party conference scheduling method based on big data optimization provided by the present invention achieves full-link intelligent collaborative optimization in the multi-party conference scheduling process by constructing a scheduling state vector model, a cross-schedulable interval screening mechanism, and a multi-objective optimization scoring system, with significant technical effects and practical value. First, by fusing and vectorizing multi-source data, the scheduling capabilities of participants are comprehensively characterized from four dimensions: time availability, spatial accessibility, resource compatibility, and time preference. This gives the scheduling input structured and behavioral dynamic characteristics, significantly improving the accuracy of expressing complex participant behaviors. Second, by introducing a Gaussian preference kernel, a spatial conflict penalty function, and a resource exclusion suppression term, a cross-schedulable interval model is constructed based on conflict perception, enabling flexible adaptation of scheduling results under multiple constraints such as time, space, and resources, and accurately distinguishing between priority scheduling and flexible scheduling through hierarchical thresholds. Finally, by designing a joint objective function that includes time matching, preference deviation, spatial cost, and schedule fragmentation indicators, supplemented by a minimum disturbance feedback mechanism, the system can quickly respond to user conflict feedback while maintaining the stability of most scheduling solutions, achieving a coordinated improvement in scheduling quality and efficiency. The synergistic effect of the above steps ultimately significantly enhanced the intelligence, adaptability and operability of conference scheduling, effectively solving problems such as low scheduling efficiency, strong response rigidity and poor user acceptance of the existing system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 This is an overall flow chart of a multi-party conference scheduling method based on big data optimization provided by the first embodiment of the present invention.

[0022] Figure 2 This is an algorithm flow chart of a multi-party conference scheduling method based on big data optimization provided in the first embodiment of the present invention.

[0023] Figure 3 This is an overall flow chart of a multi-party conference scheduling system based on big data optimization provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0024] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0025] Example 1, with reference to Figure 1 - Figure 2 , as one embodiment of the present invention, provides a multi-party conference scheduling method based on big data optimization, comprising:

[0026] S1: Collect multi-source data, perform unified time normalization and spatial geocoding on the multi-source data, and construct a scheduling state vector model, comprehensively considering the four dimensions of time availability, spatial accessibility, resource compatibility, and time preference.

[0027] Furthermore, the system acquires raw meeting-related data, including but not limited to each participant's historical meeting records, schedules, office locations, commuting radius, device usage habits, and preferred meeting time periods. Data fusion algorithms integrate this data across disparate systems. Using standardized time representation and geographic information encoding, a scheduling state vector model is constructed for each participant, characterizing their availability level for meetings within a specific time period and forming a preliminary set of schedulable window intervals. This process incorporates a dynamic time window sliding mechanism and a reachability scoring function, providing high-resolution input for subsequent matching.

[0028] Let the scheduling state vector of the i-th participant be expressed as:

[0029] S i =[T i ,L i ,R i ,P i ]

[0030] Among them, T i represents the time availability vector, L i represents the spatial accessibility vector, R i Represents the resource compatibility vector, P i represents the time preference weight vector, S i Let be the scheduling state vector of the i-th participant.

[0031] Time availability vector T i ∈{0,1} n , assuming that daily time is divided into n discrete time units (e.g., 15 minutes as the granularity, then n = 96), define:

[0032]

[0033] but:

[0034] T i =[T i,1 ,T i,2 ,…,T i,n ]

[0035] Spatial accessibility vector L i ∈[0,1] m , let the candidate set of meeting places be {l1,l2,…,l m}, define user i from current location to location l j The reachability rating is:

[0036]

[0037] Among them, d i,j Indicates that user i goes to location l j δ represents the travel tolerance scale (controlling the accessibility decay rate).

[0038] but:

[0039] L i =[L i,1 ,L i,2 ,…,L i,m ]

[0040] Resource compatibility vector R i ∈{0,1} k , suppose the resource set required for the meeting is {r1,r2,…,r k}, if user i can access resource r j ,but:

[0041]

[0042] Preference weight vector P i ∈[0,1] n Indicates the user's preference at each time unit, which can be generated based on historical behavior (attendance rate, selection rate, response time, etc.):

[0043]

[0044] Or based on a normalized Gaussian preference distribution:

[0045]

[0046] Among them, μ i is the time center when users most frequently participate in meetings, σ i is the preferred time width.

[0047] In multi-party scheduling, any set of users U = {i1,i2,…,i q}, the weighted superposition and intersection of the availability vectors are expressed as:

[0048]

[0049] in, Represents the joint scheduling feasibility score of all users at that time point.

[0050] S2: Based on natural language processing technology, it deeply analyzes the meeting scheduling intention, extracts the meeting topic, participant list, meeting duration, resource requirements and time expectations from free text, compares the meeting topic with the semantic features in historical meeting data through a semantic matching mechanism, automatically infers the resource allocation and common time preferences of the meeting, and combines it with the scheduling state vector model to output a set of candidate meeting scheduling solutions.

[0051] Furthermore, after a meeting scheduling request is submitted, the system conducts a deep analysis of the scheduling intent. Natural language processing technology is used to identify the implicit intent features contained in the scheduling request, enabling dynamic modeling of meeting requirements. Structured fields are extracted from the user-submitted request, including meeting topic keywords, meeting duration, meeting format (e.g., online, offline, or hybrid), a list of required participants, the desired time slot and its priority, and explicit requirements for auxiliary resources (e.g., projectors, screen sharing devices, interactive whiteboards, etc.).

[0052] A semantic matching mechanism is introduced to perform word embedding conversion on conference topic keywords. Combined with the semantic distribution characteristics in historical conference data, historical conference categories are matched through semantic similarity calculation to infer the typical resource demand configuration and regular time distribution pattern of such conferences.

[0053] For example, if historical "project review" meetings are typically held on Friday afternoons and prefer screen sharing and video conferencing equipment, the current request will automatically inherit this behavioral preference as a recommended parameter. This process maps the semantic understanding results into a multidimensional feature space, forming a three-dimensional set of meeting requirement parameters, including but not limited to: preferred time periods, resource allocation priorities, meeting importance weights, and spatial mode preferences (co-located collaboration vs. remote connections).

[0054] The demand parameter set will serve as the prior input for the subsequent conference scheduling modeling process. It will no longer be a static constraint, but a behavioral driver that actively guides the scheduling optimization direction. By integrating semantic information with historical behavior data to establish a dynamic mapping relationship,

[0055] It should be noted that, assuming the set of candidate meeting time periods is The set of participants is U = {1,2,…,q}, then the final set of cross-schedulable intervals is Defined as:

[0056]

[0057] Among them, T i,j represents the time availability from the scheduling state vector. If user i is at t j If it is schedulable, it is 1, otherwise it is 0. i,j represents the temporal preference score, which comes from the Gaussian preference kernel or historical statistical probability sampled in the state vector, ψ i (j) represents the spatial conflict suppression function, which is expressed as:

[0058]

[0059] Among them, d i,h is the distance from user i to the candidate meeting location h, δ i is the spatial reachability tolerance, θ i (j) represents the resource conflict suppression function, which is represented by i:

[0060]

[0061] Among them, R i,r The resource availability from the original state vector, if it is 1, it means the resource is available, ω i,r (j) represents the occupancy rate of the resource in other conferences during time period j (dynamically counted by the scheduling system), λ represents the joint evaluation threshold (e.g., 0.65), which represents the lower limit of "acceptable scheduling", is the final available time period set, t j is the jth discrete time unit, q is the total number of participants, T i,j Indicates that the i-th participant is in time period t j Is it schedulable? i,j For the time period t j The preference score of i (j) is the spatial conflict buffer function, which is used to evaluate its j The geographic attendance probability under i (j) is the resource conflict buffer function, which is used to reflect the j Equipment availability risk under i and σ i Its historical preference time center and distribution variance, ω i,r (j) is the conflict degree of the i-th user for resource r in time period j, δ i is the spatial reachability tolerance, and λ is the joint scheduling decision threshold.

[0062] When the internal score is lower than λ, the time period is considered unschedulable; between λ and 0.85 is the flexible scheduling range; and above 0.85 is considered the priority scheduling time period. The output set can be used by the scheduling optimizer to further match resource allocation and time sorting.

[0063] The multi-objective joint function is expressed as:

[0064]

[0065] in, is the candidate time period t j and the comprehensive scheduling score under the candidate location h, q is the number of participants, β i (t j ) is the i-th participant in time period t j The availability of the product of the preference score, Γ i (t j ) is the Gaussian preference response function in this time period, Ω i (h) is the spatial accessibility function to the candidate meeting location h, Φ i (t j ) is the fragmentation disturbance intensity function in this time period, μ i The most frequent meeting time center in its history, σ i Its preference distribution variance, d i,h is the distance from its current location to the meeting place, δ i is the spatial reachability tolerance coefficient, τ i (t j ) is the number of interruptions to the original schedule caused by the time period, χ i (t j ) is the frequency or probability of failure of the time period in its scheduling history, α is the exponent of the preference deviation penalty factor, and ∈ is a small constant to prevent the denominator from being zero.

[0066] like It indicates high-priority scheduling;

[0067] like The value of [0.5, 0.8) indicates a flexible scheduling area;

[0068] like If the value of is [0,0.5), it means that it is not recommended to schedule a meeting.

[0069] S3: A multi-objective scoring mechanism is introduced to evaluate the scheduling time period and push the candidate meeting scheduling solutions to the user in the form of a structured summary for confirmation. If there are conflicts in user feedback, the conflicting parts are adjusted based on the minimum disturbance optimization mechanism.

[0070] Furthermore, after solving the multi-objective integrated scheduling model, all candidate scheduling solutions are quantitatively evaluated and prioritized. Each scheduling candidate is scored based on the previously calculated comprehensive objective function and normalized into a comparable satisfaction index. This index also incorporates multiple evaluation factors such as the degree of time overlap between participants, the degree of matching of individual time preferences, spatial accessibility, and the rationality of matching auxiliary resource allocation, ensuring that the ranking results fully reflect the overall advantages and disadvantages of the scheduling solution.

[0071] To enhance user interaction and decision-making efficiency, a user-guided response mechanism has been introduced. After the scheduling engine ranks candidate solutions, the system automatically selects several optimal scheduling solutions with scores within a threshold range (e.g., the top 5% or 10%) and pushes them to the meeting initiator for initial confirmation and selection. When pushing solutions, the scheduling system also displays a summary of key parameters for each solution (e.g., scheduled time period, participant availability, meeting resource allocation, potential conflict risks, etc.), guiding users to make refined decisions while balancing efficiency and preferences.

[0072] If the meeting initiator or key participants raise conflicting feedback regarding the recommended solution, the minimum disturbance callback algorithm is triggered without global recalculation. Based on the original scheduling vector and key conflict points, the algorithm automatically locks in the arrangements of the majority of confirmed participants and, within the feasible range, fine-tunes and compensates for the conflicting local parameters (such as individual time units, resource utilization, and candidate meeting locations). This mechanism ensures the stability of the overall scheduling solution while enabling adjustments to key individual components, significantly improving scheduling response efficiency and user satisfaction.

[0073] Example 2, an embodiment of the present invention, provides a multi-party conference scheduling method based on big data optimization. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0074] First, in this experiment, to verify the effectiveness of the proposed scheduling state vector modeling mechanism and multi-factor scoring function in improving the accuracy and rationality of multi-party conference scheduling, a dataset of seven simulated participants (A1-A7) with typical differences was constructed. These participants were assigned characteristic parameters such as historical behavior, travel distance, resource matching, and time preferences. The experiment set the daily scheduling time span to 15 minutes, and each user evaluated the schedulability of a candidate time period.

[0075] The experiment first collected raw scheduling data from each participant, including whether they were available in the current time period (time availability T), their physical distance from the meeting location (used to calculate spatial accessibility L), whether they had the necessary equipment for the current meeting (resource compatibility R), and their historical preferences for attending meetings in that time period (time preference P). These data were uniformly encoded into a scheduling state vector structure. Spatial accessibility was scored using an exponential decay function (using a distance of 2 km as the decay baseline), while time preference was calculated based on the success rate of behavioral statistics or converted into a standard Gaussian kernel response function.

[0076] The validity score for this time period was then calculated using both a traditional method (considering only T) and the proposed method (a comprehensive interaction function of T, L, R, and P). The traditional method uses "idleness" as the criterion for a 0-1 judgment or a basic score based on the multiplication of T and P. The proposed method further integrates spatial attenuation factors and resource availability factors, introduces a preference deviation attenuation mechanism, and introduces a fragmentation control term to nonlinearly modulate the scheduling score, thereby more accurately reflecting the comprehensive scheduling feasibility of participants in the current time period.

[0077] Table 1 Experimental data table

[0078]

[0079] In this experiment, to verify the effectiveness of the proposed scheduling state vector modeling mechanism and multi-factor scoring function in improving the accuracy and rationality of multi-party conference scheduling, a dataset of seven simulated participants (A1-A7) with typical differences was constructed. These participants were assigned characteristic parameters such as historical behavior, travel distance, resource matching, and time preferences. The experiment set the daily scheduling time span to 15 minutes, and each user evaluated the schedulability of a candidate time period.

[0080] The experiment first collected raw scheduling data from each participant, including whether they were available in the current time period (time availability T), their physical distance from the meeting location (used to calculate spatial accessibility L), whether they had the necessary equipment for the current meeting (resource compatibility R), and their historical preferences for attending meetings in that time period (time preference P). These data were uniformly encoded into a scheduling state vector structure. Spatial accessibility was scored using an exponential decay function (using a distance of 2 km as the decay baseline), while time preference was calculated based on the success rate of behavioral statistics or converted into a standard Gaussian kernel response function.

[0081] The validity score for this time period was then calculated using both a traditional method (considering only T) and the proposed method (a comprehensive interaction function of T, L, R, and P). The traditional method uses "idleness" as the criterion for a 0-1 judgment or a basic score based on the multiplication of T and P. The proposed method further integrates spatial attenuation factors and resource availability factors, introduces a preference deviation attenuation mechanism, and introduces a fragmentation control term to nonlinearly modulate the scheduling score, thereby more accurately reflecting the comprehensive scheduling feasibility of participants in the current time period.

[0082] Example 3, reference Figure 3 , is an embodiment of the present invention, which provides a multi-party conference scheduling system based on big data optimization, including a scheduling status modeling and demand semantic parsing module, a cross-availability interval construction and multi-objective scheduling optimization module, and a scheduling result sorting feedback and local callback correction module.

[0083] The scheduling status modeling and demand semantics analysis module is used to extract the input variables required for scheduling and establish behavioral capability portraits of the participants. The cross-available interval construction and multi-objective scheduling optimization module is used to find the optimal meeting arrangement time period and location configuration that meets the participants, resource constraints, and preference goals. The scheduling result sorting feedback and local callback correction module is used to sort according to the comprehensive satisfaction index and recommend it to the meeting initiator for selection through a guided interactive interface.

[0084] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0085] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0086] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0087] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logical functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc. It should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention and are not intended to be limiting. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced with equivalents without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications should be encompassed by the claims of the present invention.

[0088] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multi-party conference scheduling method based on big data optimization, characterized in that: include: Collect multi-source data, perform unified time normalization and spatial geocoding on the multi-source data, and build a scheduling state vector model that comprehensively considers the four dimensions of time availability, spatial accessibility, resource compatibility, and time preference; Using natural language processing technology, the system deeply analyzes meeting scheduling intentions, extracting the meeting topic, attendee list, meeting duration, resource requirements, and time expectations from free text. Using a semantic matching mechanism, the system compares the meeting topic with semantic features in historical meeting data, automatically inferring the meeting's resource allocation and common time preferences. Combined with the scheduling state vector model, the system outputs a set of candidate meeting scheduling solutions. A multi-objective scoring mechanism is introduced to evaluate the scheduling time period, and the candidate meeting scheduling plans are pushed to the user for confirmation in the form of a structured summary. If there is a conflict in user feedback, the conflicting part is adjusted based on the minimum disturbance optimization mechanism.

2. The multi-party conference scheduling method based on big data optimization according to claim 1, characterized in that: The multi-source data includes the user's historical meeting records, personal schedule, office location and commuting range, conference equipment usage habits and time preference information.

3. The multi-party conference scheduling method based on big data optimization according to claim 2, characterized in that: The semantic matching mechanism includes using word embedding conversion technology to vectorize the conference topic keywords during the semantic parsing process, and combining the semantic distribution characteristics of historical meetings to perform similarity analysis between the current scheduling request and existing conference types, and infer the resource requirements and time pattern of the meeting.

4. The multi-party conference scheduling method based on big data optimization according to claim 3, characterized in that: The outputting of the conference scheduling candidate solution set includes setting a conflict detection mechanism during the generation of the scheduling candidate solution, including time conflict, space conflict and resource conflict, introducing a flexible scheduling boundary strategy, and performing tolerance judgment and dynamic intervention.

5. The multi-party conference scheduling method based on big data optimization according to claim 4, characterized in that: The evaluation of the scheduling time period includes dynamically weighting the priorities of the candidate time periods in combination with the historical behavior data of the user.

6. The multi-party conference scheduling method based on big data optimization according to claim 5, characterized in that: The multi-objective scoring mechanism includes time period overlap rate, personal time preference, spatial accessibility, meeting resource availability and schedule fragmentation degree evaluation indicators, and performs weighted sorting and priority output on all candidate solutions.

7. The multi-party conference scheduling method based on big data optimization according to claim 6, characterized in that: The method of pushing the candidate conference scheduling solutions to the user for confirmation in the form of a structured summary includes adopting a minimum disturbance optimization mechanism to adjust the time period or resource configuration when the user provides conflicting feedback on the candidate scheduling solutions, and retaining the confirmed participant arrangements and resource allocations.

8. A system using the multi-party conference scheduling method based on big data optimization according to any one of claims 1 to 7, characterized in that: It includes scheduling status modeling and demand semantic analysis modules, cross-availability interval construction and multi-objective scheduling optimization modules, scheduling result sorting feedback and local callback correction modules; The scheduling state modeling and demand semantic parsing module is used to extract the input variables required for scheduling and establish a behavioral capability profile of the participants; The cross-availability interval construction and multi-objective scheduling optimization module is used to find the optimal meeting schedule time and location configuration that meets the participants, resource constraints, and preference objectives; The scheduling result sorting feedback and local callback correction module is used to sort according to the comprehensive satisfaction index and recommend it to the conference initiator for selection through a guided interactive interface.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the multi-party conference scheduling method based on big data optimization according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-party conference scheduling method based on big data optimization according to any one of claims 1 to 7 are implemented.

Images