CN-121806558-B - Exhibition hall scene self-adaptive intelligent switching control system and method

CN121806558BCN 121806558 BCN121806558 BCN 121806558BCN-121806558-B

Abstract

The invention belongs to the technical field of exhibition hall control, and particularly discloses an exhibition hall scene self-adaptive intelligent switching control system and method. By acquiring identity background, declaration preference and real-time behavior data, dynamically generating individual interest vectors and identifying real-time intent states via an individual vector building module, and calculating the individual-group matching degree by using an individual deviation analysis module, and finally realizing intelligent matching of three-level service modes of group navigation, slight individuation and deep individuation by using a service mode determination module. The invention realizes the technical span from static prejudgment to real-time perception, establishes a quantitative matching mechanism of individual and group interests, solves the difficult problem of service mode grading decision, and effectively improves the intelligent level of exhibition hall service and the satisfaction degree of visitors.

Inventors

CHANG LU
ZHAO QING
WANG XU
YAN PING

Assignees

山东百特展览工程有限公司

Dates

Publication Date: 20260508
Application Date: 20260305

Claims (9)

1. An exhibition hall scene self-adaptive intelligent switching control system, which is characterized by comprising: The visitor information acquisition module is used for acquiring visitor information, Collecting identity background, declaration preference data, real-time behavior data and position information of a visitor, and generating a visitor information data set; The individual vector establishing module is used for outputting dynamically evolving individual interest vectors based on the data set and the time sequence analysis of the behavior data, and specifically, generating initial interest intensity values based on statement preference data, form types and historical visit records; calculating behavior interest intensity based on the statistical normalization residence time length and normalization interaction frequency of the time sequence behavior data of the visitor, linearly weighting and fusing the behavior interest intensity and the initial interest intensity value to generate an updated interest intensity value, and outputting individual interest vectors which dynamically evolve based on all preference categories and the updated interest intensity value; the identifying the real-time intent state of the visitor includes: Acquiring a real-time behavior data stream of a visitor; Counting total duration of stay before setting the display item, average moving speed, stay proportion of the sight on the display item, operation interval time of each interaction device and tortuosity of the moving path based on the behavior data stream; Comparing the counted characteristic indexes with a preset intention state threshold value, and outputting a real-time intention state, wherein the intention state comprises at least one of deep visit, quick pass, exploration searching and interaction experience; Storing the state identification in association with the timestamp; the group matching analysis module dynamically clusters the visitors according to the real-time positions and the individual interest vectors, generates group preference vectors, calculates the similarity between the individual interest vectors of each visitor and the corresponding group preference vectors through cosine similarity, and generates individual-group matching degree; The service mode determining module is used for comparing the individual-group matching degree with a preset threshold value and matching a preset service mode according to a comparison result; and the control instruction generation execution terminal generates a corresponding scene control instruction based on the matched service mode and distributes the corresponding scene control instruction to the corresponding exhibition area equipment controller to execute scene switching control.
2. The system of claim 1, wherein the system comprises: the individual interest vector output includes: extracting free text or selected preference tags input by a user from declarative preference data; carrying out semantic similarity calculation on the free text or the preference label and semantic expansion sets of all preset preference categories, and mapping the preference categories according to calculation results; Assigning an initial interest intensity value to each selected preference label according to a preset assignment rule; Inquiring a historical visit record of a visitor, if an intersection exists between a historical effective preference set and a current declaration preference, multiplying an initial interest intensity value of a preference category in the intersection by a preset strengthening coefficient, and outputting a final initial interest intensity value; Periodically acquiring and quantifying time sequence behavior data of visitors through a sensor network deployed in an exhibition area, and counting normalized residence time under each preference category associated exhibition item and normalized interaction frequency of an associated interaction device; For each preference category, calculating the action interest intensity of the current update period by linearly weighting and summing the normalized residence time length and the normalized interaction frequency; Performing linear weighted fusion on the behavioral interest intensity and the initial interest intensity value of the corresponding preference category to generate an updated interest intensity value; Dynamically evolving individual interest vectors are generated based on all preference categories and corresponding updated interest intensity values.
3. The system for controlling the adaptive intelligent switching of the exhibition hall according to claim 2, wherein the specific content of the preset allocation rule comprises: Extracting the form type to which the preference label belongs; if the preference label is from a single menu, the selected unique preference label obtains the preset highest initial interest intensity value ; If the preference labels are derived from a multi-choice form, assigning the same initial interest intensity value to each selected preference label based on the preset highest initial interest intensity value; if the preference tag is derived from a rating questionnaire, the following steps are performed: The method comprises the steps of obtaining a grade identifier selected by a visitor and a highest grade identifier, and calculating the number of interval grades between the grade identifier and the highest grade identifier; Calculating the complement of the ratio of the number of interval grades to the total number of grades to obtain the relative position proportion; calculating an initial interest intensity value by a preset nonlinear mapping function , , For the preset lowest initial interest intensity value, Is a configured sensitivity coefficient.
4. The system of claim 2, wherein the system comprises: the mapping process of the preference category comprises the following steps: judging whether the similarity score calculated corresponding to the preset preference category is higher than a preset matching threshold value or not; if not, marking the declaration preference as an unidentified item and storing the unidentified item into a temporary interest pool; If so, judging whether the number of preference categories with similarity scores exceeding a matching threshold is 1; if the similarity score is 1, mapping the declaration preference to the category, otherwise, taking the preset preference category with the highest similarity score as a main mapping category, taking the rest as a secondary mapping category, and carrying out corresponding marking.
5. The system for adaptively and intelligently switching and controlling the exhibition hall scene according to claim 4, wherein said calculation of the strength of interest of the behavior further comprises the steps of matching weight combinations according to preference categories, and specifically comprises the following steps: If the preference category calculated at present is the main mapping, adopting a first weight combination, wherein the normalized residence time length weight is greater than the normalized interaction frequency weight; and if the mapping is sub-mapping, adopting a second weight combination, wherein the normalized residence time length weight is smaller than the normalized interaction frequency weight.
6. The system of claim 1, wherein the specific generation of the group preference vector comprises: dynamically grouping visitors based on the position coordinates through a DBSCAN clustering algorithm to form a temporary guide group; Extracting individual interest vectors of all members in each group, and calculating the average value and variance of the interest intensity of the same preference category in the group; setting a member weight initial value based on the interest intensity average value and the variance, and judging whether the member triggers the following conditions: the method comprises the following steps that 1, the class interest intensity value in a member is lower than the group average value and the variance is higher than a discrete threshold value; the residence time of the member at the current position exceeds the average residence time of the group; Condition 3, the member is identified as a deep visit status; if the triggering condition 1 is met, reducing the initial value of the member weight to a first weight value; if the condition 2 is triggered, the initial value of the member weight is lifted to a second weight value; if the triggering condition 3 is met, the initial value of the member weight is raised to a third weight value, and the third weight value is larger than the second weight value; If the condition is not triggered, keeping the initial value of the member weight unchanged; Performing linear weighted fusion calculation on the final member weight value and the interest intensity value, and dividing the final member weight value by the total number of the group members to obtain a final group interest intensity value; the final population interest intensity values for all preference categories will be combined to generate a population preference vector.
7. The system for adaptively and intelligently switching and controlling the exhibition hall according to claim 1, wherein the matching process of the preset service mode comprises the following steps: when the individual-population matching degree is greater than or equal to a first threshold, matching the population navigation mode; matching the light personalization mode when the individual-population match is between a first threshold and a second threshold, wherein the first threshold is less than the second threshold; When the individual-population matching degree is smaller than the second threshold, matching the depth personalized pattern.
8. The system of claim 7, wherein the generating of the scene control instruction comprises: if the group guide mode is adopted, searching a matched scene configuration scheme from a scene template library based on the group preference vector, and generating a corresponding integrated control instruction; if the model is a mild individuation model, extracting differentiated preference dimensions in individual interest vectors on the basis of the integrated control instruction to generate a supplementary scene instruction; If the depth individuation mode is adopted, generating a scene switching sequence instruction containing an exclusive visit path based on the individual interest vector, and preloading individuation configuration parameters for relevant exhibition area equipment to execute the instruction.
9. A self-adaptive intelligent switching control method for a exhibition hall scene is characterized by comprising the following steps: collecting identity background, declaration preference data, real-time behavior data and position information of a visitor, and generating a visitor information data set; based on the data set, the time sequence analysis of the behavior data is combined to output a dynamically evolved individual interest vector, and the method comprises the steps of generating an initial interest intensity value based on statement preference data, form types and historical visit records, calculating the behavior interest intensity based on the statistical normalization residence time length and normalization interaction frequency of the visitor time sequence behavior data, linearly weighting and fusing the behavior interest intensity and the initial interest intensity value to generate an updated interest intensity value, outputting the dynamically evolved individual interest vector based on all preference types and the updated interest intensity value, and identifying the real-time intention state of the visitor; the identifying the real-time intent state of the visitor includes: Acquiring a real-time behavior data stream of a visitor; Counting total duration of stay before setting the display item, average moving speed, stay proportion of the sight on the display item, operation interval time of each interaction device and tortuosity of the moving path based on the behavior data stream; Comparing the counted characteristic indexes with a preset intention state threshold value, and outputting a real-time intention state, wherein the intention state comprises at least one of deep visit, quick pass, exploration searching and interaction experience; Storing the state identification in association with the timestamp; Dynamically clustering visitors to generate group preference vectors, calculating the similarity between the individual interest vector of each visitor and the corresponding group preference vector through cosine similarity, and generating individual-group matching degree; comparing the individual-group matching degree with a preset threshold value, and matching a preset service mode according to a comparison result; And generating a corresponding scene control instruction based on the matched service mode, and distributing the corresponding scene control instruction to a corresponding exhibition area equipment controller to execute scene switching control.

Description

Exhibition hall scene self-adaptive intelligent switching control system and method Technical Field The invention belongs to the technical field of exhibition hall control, and particularly relates to an exhibition hall scene self-adaptive intelligent switching control system and method. Background The exhibition hall is taken as an important place for cultural exhibition and knowledge transmission, and along with the development of digital technology, intelligent scene control not only relates to the optimization of the exhibition effect, but also directly influences the immersion and satisfaction of visitors, thereby highlighting the importance of control management of the exhibition hall. The method, the device, the equipment and the storage medium for controlling the self-adaptive switching of the exhibition hall scene disclosed in the China patent application with the application number 202411764637.8 in the prior art generate a waiting queue and personal interestingness ranking aiming at each exhibition area by integrating check-in, reservation time and personal association information of visitors, so that interest matching and queuing management are combined, the same team of interest players is realized, and the balance of individuation and efficiency is maintained. When the prior art realizes the self-adaptive switching of the exhibition hall scene, the exhibition hall scene mainly follows a paradigm centered on scheduling, and visitors are regarded as schedulable units to realize the optimal macroscopic efficiency. However, in the process of pursuing the balance between group efficiency and overall experience, the paradigm has the following fundamental defects that 1, prior information such as reservation, sign-in and the like before visitor entering a garden is relied on, dynamic behavior data generated in the visiting process is lacking, navigation content or scene presentation cannot be dynamically adjusted according to real-time interest change and emotion feedback of visitor, and the overall navigation process experience is stiff. 2. The credibility and timeliness of the interestingness ranking generated only by the prior information are difficult to ensure, and historical visit records, real-time interaction data and the like are not comprehensively utilized for cross verification, so that the accuracy of interest assessment and matching is inaccurate. 3. The maximization of the commonality interest of the existing target group is completely free from consideration of the intensity of the special preference of the individual visitor and the urgent need of the individual visitor for personalized services. Without establishing a per-situation management mechanism, it is not possible to distinguish which guests are suitable for group navigation and which guests need more personalized targeted services. Disclosure of Invention In view of this, in order to solve the above-mentioned problems, an exhibition hall scene adaptive intelligent switching control system and method are proposed. The invention provides an exhibition hall scene self-adaptive intelligent switching control system, which comprises a visitor information acquisition module, a visitor information data set and a visitor information data set, wherein the visitor information acquisition module is used for acquiring identity background, declaration preference data, real-time behavior data and position information of a visitor. And the individual vector establishing module is used for outputting dynamically evolving individual interest vectors based on the data set and combining time sequence analysis of the behavior data, and identifying the real-time intention state of the visitor. The group matching analysis module dynamically clusters the visitors according to the real-time positions and the individual interest vectors, generates group preference vectors, calculates the similarity between the individual interest vectors of each visitor and the corresponding group preference vectors through cosine similarity, and generates individual-group matching degree. And the service mode determining module is used for comparing the individual-group matching degree with a preset threshold value and matching a preset service mode according to a comparison result. And the control instruction generation execution terminal generates a corresponding scene control instruction based on the matched service mode and distributes the corresponding scene control instruction to the corresponding exhibition area equipment controller to execute scene switching control. The invention also provides a system method for controlling the self-adaptive intelligent switching of the exhibition hall scene, which comprises the steps of collecting identity background, statement preference data, real-time behavior data and position information of the visitor and generating a visitor information data set. Based on the data set, the individual interest vectors which dynamic