CN-121981805-A - Product popularization method, system and equipment based on cross-source time sequence logic map

Abstract

The embodiment of the specification discloses a product popularization method, a system and equipment based on a cross-source time sequence logic map, wherein the product popularization method comprises the steps of obtaining text content, text interaction behavior data and text release time in a community platform, processing to obtain a standardized dataset which can be used for semantic analysis, carrying out semantic analysis based on the text content, obtaining hot topics in the community platform, mapping the hot topics to a product to be promoted to obtain an optimal popularization cutting-in angle, constructing a logic map based on the hot topics, the optimal popularization cutting-in angle, the text interaction behavior data and the text release time, obtaining a popularization response trend based on the logic map and combining event influence paths, group emotion evolution trends and time attenuation factors, and generating and releasing a popularization document matched with the hot topics based on the optimal popularization cutting-in angle and the popularization response trend. And the intelligent and self-adaptive optimization of popularization opportunity, content and strategy is realized.

Inventors

• JI BAIYANG
• CHEN ZHANGLI

Assignees

• 浙江工业大学

Dates

Publication Date

20260505

Application Date

20260130

Claims (10)

1. 1. The product popularization method based on the cross-source time sequence logic map is characterized by comprising the following steps of: acquiring text content, text interaction behavior data and text release time in a community platform, and processing to obtain a standardized data set for semantic analysis; semantic analysis is carried out based on text content in the standardized dataset, and hot topics in a community platform are obtained; Mapping the hot topic semantics to the product to be promoted so as to obtain the optimal promotion cutting-in angle of the product to be promoted; constructing logic maps based on hot topics, optimal popularization cut-in angles, text interaction behavior data in a standardized dataset and text release time; Acquiring a popularization response trend based on logic maps and combining event influence paths, group emotion evolution trend and time attenuation factors; and generating and issuing a promotion document matching the hot topic based on the optimal promotion cutting-in angle and the promotion response trend.
2. 2. The method for product popularization based on cross-source time sequence logic map according to claim 1, wherein the semantic analysis is performed on text content in the standardized dataset to obtain hot topics in a community platform, comprising: coding text contents in the standardized data set by adopting a semantic representation model to obtain semantic vector representations corresponding to each text; Acquiring weights of each semantic vector representation based on text interaction behavior data and text release time in a standardized data set; And aggregating based on the weight of each semantic vector representation and the similarity relation among each semantic vector representation to obtain hot topics and semantic vector representations thereof in the community platform.
3. 3. The method for product promotion based on cross-source time sequence logic map according to claim 2, wherein the mapping the hot topic semantics to the product to be promoted to obtain the optimal promotion cut-in angle of the product to be promoted comprises: Acquiring concept unit sets which comprise semantic vector representations respectively corresponding to products to be promoted when different promotion cutting-in angles express the products to be promoted; And carrying out similarity calculation on semantic vector representation based on hot topics and the concept unit set, and taking the popularization cut-in angle with the highest similarity as the optimal popularization cut-in angle of the product to be popularized.
4. 4. The method for product popularization based on cross-source time sequence logic map according to claim 3, wherein the constructing logic map based on the hot topic, the optimal popularization cut-in angle and text interaction behavior data and text release time in the standardized dataset includes: acquiring hot related information from official text data on an official platform based on the hot topic; acquiring logic external event nodes of the map based on the hot topics and the hot related information; Acquiring internal popularization concept nodes of logic maps based on the optimal popularization cut-in angle; Acquiring group emotion state nodes of logic maps based on text interaction behavior data in a standardized dataset; based on the external event nodes, the internal popularization concept nodes, the group emotion state nodes and the text release time in the standardized dataset, logic association relations and time constraint and evolution relations among the nodes are established to obtain a logic map.
5. 5. The method for product popularization based on cross-source time sequence logic map as claimed in claim 4, wherein the establishing logic association relationship and time constraint and evolution relationship between nodes based on external event nodes, internal popularization concept nodes, group emotion state nodes and text release time in standardized dataset includes: obtaining an objective emotion state vector corresponding to the external event node and representing objective emotion distribution; Acquiring an external emotion impact vector corresponding to the current time window based on the objective emotion state vector corresponding to each external event node in the current time window, and obtaining logic association relations; And acquiring the change condition of the group emotion state nodes corresponding to the internal popularization concept nodes in different time windows based on the text release time in the standardized data set, and obtaining the time constraint and evolution relation.
6. 6. The method for promoting products based on a cross-source time sequence logic map, which is characterized in that the group emotion state nodes comprise group emotion state vectors representing group emotion distribution, the promotion response trend is obtained based on the logic map by combining event influence paths, group emotion evolution trend and time attenuation factors, and the method comprises the following steps: calculating the path influence force corresponding to each event influence path from the external event node to the internal popularization concept node through the group emotion state node based on logic patterns, and obtaining the maximum influence force path; Acquiring a group emotion state vector corresponding to the next time window based on an external emotion impact vector corresponding to the current time window and a group emotion state vector of a group emotion state node so as to represent a group emotion evolution trend; and calculating a promotion adaptation index based on the influence maximum path and the population emotion evolution trend so as to represent a promotion response trend.
7. 7. The method for product popularization based on the cross-source time sequence logic map according to claim 6, wherein the obtaining the external emotion impact vector corresponding to the current time window based on the objective emotion state vector corresponding to each external event node in the current time window includes: acquiring an external emotion impact vector corresponding to the current time window based on the objective emotion state vector corresponding to each external event node in the current time window and the influence weight of the external emotion state vector; The product popularization method further comprises the following steps: acquiring popularization effect data of a popularization document; and updating the respective corresponding influence weight of each external event node based on the promotion effect data.
8. 8. The product popularization method based on the cross-source time sequence logic map as claimed in claim 7, wherein the generating and issuing the popularization document matching the hot topic based on the optimal popularization cut-in angle and the popularization response trend includes: Based on the historical popularization text and corresponding popularization effect data, a deep learning generation model is adopted to acquire a plurality of candidate strategy rules, each candidate strategy rule is quantitatively evaluated, and candidate strategy rules qualified in evaluation are added into a strategy rule library; and selecting an optimal popularization strategy from a strategy rule base based on the optimal popularization cut-in angle and the popularization response trend, and generating and releasing a popularization document matched with the hot topic by adopting the optimal popularization strategy.
9. 9. The product popularization system based on the cross-source time sequence logic map is characterized by comprising an information acquisition and processing unit, a semantic analysis unit, a semantic matching unit, a map construction unit, a popularization trend prediction unit and a popularization decision and execution unit: The information acquisition and processing unit acquires text content, text interaction behavior data and text release time in the community platform and processes the text content, the text interaction behavior data and the text release time to obtain a standardized data set for semantic analysis; the semantic analysis unit performs semantic analysis based on text content in the standardized dataset to acquire hot topics in the community platform; The semantic matching unit is used for mapping the hot topic semantic to the product to be promoted so as to acquire the optimal promotion cutting-in angle of the product to be promoted; the map construction unit is used for constructing logic maps based on hot topics, optimal popularization cut-in angles, text interaction behavior data in a standardized dataset and text release time; The promotion trend prediction unit is used for acquiring promotion response trends based on logic maps and combining event influence paths, group emotion evolution trends and time attenuation factors; And the promotion decision and execution unit generates and issues a promotion document matching the hot topic based on the optimal promotion cut-in angle and the promotion response trend.
10. 10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1-8 when the computer program is executed.

Description

Product popularization method, system and equipment based on cross-source time sequence logic map Technical Field The embodiments of the present specification relate to the field of internet marketing technology, and in particular, to optimization of product promotion accuracy. Background With the rapid development of internet finance, an investment communication platform becomes an important channel for investors to acquire information and express views. Under the background, enterprises perform popularization and operation of financial products in the investment communities, aim to reach target user groups accurately and improve user conversion efficiency, and form a continuously-growing market demand in the field. Currently, in the product popularization practice in the field, the decision mode mainly depends on manual operation or a simple automatic tool based on keyword matching. Specifically, existing systems or operators typically monitor indicators of a single or limited dimension, such as community topic popularity, keyword frequency, and the like, and combine with human experience to formulate content directions and select release opportunities. In addition, some prior art technologies also use knowledge maps or logic maps, through which structured descriptions and query searches of events and entity relationships that have occurred, exist as a static information base that assists in decision making. However, the decision mode based on static information and experience judgment causes hysteresis of popularization decisions relative to the change of market situation, and the application of the map is limited to the association of information and query retrieval. The accuracy of the real-time dynamic matching of the popularization strategy and the market is insufficient, and the popularization effect of the product is limited. Disclosure of Invention The embodiment of the specification provides a product popularization method, a system and equipment based on a cross-source time sequence logic map, which construct a cross-source time sequence logic map with predictive reasoning capability, realize predictive decision, and change a decision mode of financial product popularization from 'static response' to 'dynamic prediction', thereby realizing intelligent and self-adaptive optimization of popularization opportunity, content and strategy. The technical scheme is as follows: In a first aspect, embodiments of the present disclosure provide a product popularization method based on a cross-source timing logic map, including the following steps: acquiring text content, text interaction behavior data and text release time in a community platform, and processing to obtain a standardized data set for semantic analysis; semantic analysis is carried out based on text content in the standardized dataset, and hot topics in a community platform are obtained; Mapping the hot topic semantics to the product to be promoted so as to obtain the optimal promotion cutting-in angle of the product to be promoted; constructing logic maps based on hot topics, optimal popularization cut-in angles, text interaction behavior data in a standardized dataset and text release time; Acquiring a popularization response trend based on logic maps and combining event influence paths, group emotion evolution trend and time attenuation factors; and generating and issuing a promotion document matching the hot topic based on the optimal promotion cutting-in angle and the promotion response trend. As a preferred solution, semantic analysis is performed based on text content in a standardized dataset, and hot topics in a community platform are obtained, including: coding text contents in the standardized data set by adopting a semantic representation model to obtain semantic vector representations corresponding to each text; Acquiring weights of each semantic vector representation based on text interaction behavior data and text release time in a standardized data set; And aggregating based on the weight of each semantic vector representation and the similarity relation among each semantic vector representation to obtain hot topics and semantic vector representations thereof in the community platform. As a preferred solution, mapping hot topic semantics to a product to be promoted to obtain an optimal promotion cut-in angle of the product to be promoted, including: Acquiring concept unit sets which comprise semantic vector representations respectively corresponding to products to be promoted when different promotion cutting-in angles express the products to be promoted; And carrying out similarity calculation on semantic vector representation based on hot topics and the concept unit set, and taking the popularization cut-in angle with the highest similarity as the optimal popularization cut-in angle of the product to be popularized. As a preferred solution, constructing logic a map based on a hot topic, an optimal launch angle, and text inter