Scenic spot rockfall risk investigation method based on public observation

Abstract

The invention discloses a scenic spot rockfall risk investigation method based on public observation, belongs to the technical field of geological disaster monitoring, is used for scenic spot rockfall risk investigation, and comprises the steps of obtaining scenic spot social media text data, and performing keyword filtering on the social media text data; constructing a scenic spot rockfall risk semantic network fused with public emotion, including node construction fused with public emotion, emotion node weight calculation and scenic spot node score calculation; based on construction of a scenic spot rockfall risk semantic network fused with public emotion, scenic spots with high potential rockfall risk are analyzed and checked. Compared with the prior art, information about the potential rockfall risk of the scenic spot can be collected in a short time, monitoring of the whole scenic spot is achieved, and a monitoring blind area is avoided; the time interval from risk generation to risk discovery is shortened, and the possibility and the harm degree of rock falling accidents are effectively reduced; the economic burden of the scenic spot can be reduced, and resources are put into other safety guarantee and scenic spot construction work.

Description

Technical Field

[0001] The invention discloses a method for investigating rockfall risks in a scenic area based on public observation, and belongs to the technical field of geological disaster monitoring. Background Art

[0002] In the field of scenic area safety management, rockfall risk is an important factor that seriously threatens the safety of tourists' lives and the normal operation of scenic areas. Traditional scenic area rockfall risk investigation technology mainly relies on professional geological survey teams and professional equipment. These methods usually include regular manual on-site surveys, topographic mapping of mountains using high-precision measuring instruments (such as total stations, three-dimensional laser scanners, etc.), and obtaining data through professional monitoring equipment installed in scenic areas (such as crack monitors, microseismic sensors, etc.). However, these traditional methods have many limitations. Professional geological surveys and measurements require a lot of manpower, material resources and time costs. The frequency of regular manual on-site surveys is limited, and sudden changes in rockfall risks between two surveys may be missed. Although high-precision measuring instruments can obtain detailed mountain data, they are complicated to operate and require professional personnel to operate and interpret the data. In addition, the installation and maintenance costs of the equipment are high, and it is difficult to cover large areas of the scenic area. At the same time, the number of professional monitoring equipment is often limited, and it is impossible to effectively monitor the potential rockfall areas of the entire scenic area in an all-round way.

[0003] With the booming tourism industry, the number of tourists in scenic spots is increasing, and the safety of scenic spots is facing higher requirements. Once a rockfall accident occurs, it may cause serious casualties and property losses, and will also have a great negative impact on the reputation of the scenic spot. Therefore, scenic area managers urgently need a more efficient, comprehensive and real-time rockfall risk investigation technology to ensure the safety of tourists in the scenic area. In recent years, with the widespread popularization of mobile Internet and smart devices, the public has a stronger ability to obtain and disseminate information. When tourists are playing in scenic spots, they often observe the surrounding environment unconsciously, and their smart devices (such as mobile phones, cameras, etc.) can record various scenes in the scenic area. These large amounts of scattered observation information from the public have great potential value in rockfall risk investigation, but they have not been effectively utilized at present, and there is a lack of corresponding technology to integrate public observation into the rockfall risk investigation system of scenic spots. By integrating social media data and performing semantic analysis on it, it is possible to analyze the potential rockfall risk in scenic spots, thereby achieving early warning of potential risks in scenic spots. Summary of the invention

[0004] The purpose of the present invention is to provide a method for investigating rockfall risks in scenic spots based on public observation, so as to solve the problems in the prior art that the investigation of rockfall risks in scenic spots has high manpower, material and time costs and cannot comprehensively monitor the entire scenic spot.

[0005] A method for investigating rockfall risks in scenic spots based on public observation, comprising:

[0006] S1. Obtaining the scenic spot social media text data and filtering the social media text data by keywords;

[0007] S2. Construct a semantic network of rockfall risk in scenic spots that integrates public sentiment;

[0008] S3. Based on the construction of a semantic network of rockfall risk in scenic spots that integrates public sentiment, analyze and identify scenic spots with high potential rockfall risks.

[0009] S1 includes obtaining social media text data related to scenic spots through the Internet.

[0010] S1 performs keyword filtering on social media text data, including deleting noise data and duplicate data, and extracting keywords. The keywords include public sentiment information, scenic spot names, and location information related to rockfall risks in social media text data, and constructs a scenic spot rockfall risk analysis database.

[0011] S2 includes sentiment information extraction and keyword analysis of social media text data based on the scenic spot rockfall risk analysis database, and integrates sentiment information and scenic spot information into the semantic network by calculating the node sentiment weight.

[0012] S2 constructs a semantic network of rockfall risk in scenic spots that integrates public emotions, including node construction that integrates public emotions, emotional node weight calculation, and scenic spot node score calculation.

[0013] S2 integrates the node construction of public emotions, including identifying emotional words and scenic spot words in social media text data through keyword screening, word frequency statistics and co-occurrence relationship analysis, and constructing emotional nodes and scenic spot nodes;

[0014] Emotional vocabulary includes positive emotion and negative emotion. The extracted emotional vocabulary is used as a sub-node in the semantic network. The sub-node types include positive emotion nodes and negative emotion nodes.

[0015] Scenic spot vocabulary includes scenic spot name and location information. The extracted scenic spot vocabulary is used as the parent node in the semantic network. The parent node types include positive sentiment nodes, neutral sentiment nodes and negative sentiment nodes.

[0016] S2 emotional node weight calculation includes calculating the number of positive emotional nodes S in all child nodes connected to the parent node p , the number of negative sentiment nodes S n , the weight of positive sentiment nodes W p and the weight of negative sentiment nodes W n .

[0017] The calculation of the S2 scenic spot node score includes calculating the weight ratio of negative emotion nodes to positive emotion nodes to obtain the score P of the parent node. When 0≤P<1, the parent node type is a positive emotion node, indicating that the public's positive emotion towards the scenic spot is greater than the negative emotion, and the risk of rockfall in the scenic spot is low risk, and the risk is lower when P is closer to 0; when P=1, the parent node type is a neutral emotion node, indicating that the risk of rockfall in the scenic spot is medium risk; when P>1, the parent node type is a negative emotion node, indicating that the public's negative emotion towards the scenic spot is greater than the positive emotion, and the risk of rockfall in the scenic spot is high risk, and the larger P is, the higher the risk.

[0018] In S2, the sentiment of the parent node is analyzed, a semantic network of rockfall risk in scenic spots that integrates public sentiment is constructed, the semantic network structure is optimized, and the scenic spots discussed by the public and the corresponding semantic relations are identified in combination with scenic spot detection to obtain the potential rockfall risk of scenic spots.

[0019] S3 includes calculating the value and weight of each child node in the semantic network, evaluating the association strength between the child node and the parent node, analyzing the sentiment type of the parent node, sorting the parent node according to the negative sentiment intensity, calculating the comprehensive risk index of the scenic area, and predicting and evaluating the potential rockfall risk in the scenic area.

[0020] Compared with the prior art, the present invention has the following beneficial effects: through the real-time collection and analysis of social media data, information on potential rockfall risks in scenic spots can be collected in a short period of time, and the entire scenic spot can be monitored, effectively avoiding monitoring blind spots; the time interval from the generation of risks to their discovery is shortened, so that the scenic area management department can respond quickly and take emergency measures such as evacuating tourists and setting up warning signs in a timely manner, effectively reducing the possibility and degree of harm of rockfall accidents; it can reduce the economic burden on scenic spots and invest resources in other safety assurance and scenic area construction work. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of the technical process of the present invention;

[0022] Figure 2 This is a schematic diagram of the semantic network of rockfall risk in scenic spot A of the present invention;

[0023] Figure 3 This is a schematic diagram of the semantic network of rockfall risk in scenic area B of the present invention;

[0024] Figure 4 This is a schematic diagram of the semantic network of rockfall risk in scenic area C of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] A method for investigating rockfall risks in scenic spots based on public observation, the technical flow diagram of the present invention is as follows: Figure 1 As shown, the steps include:

[0027] S1. Obtaining the scenic spot social media text data and filtering the social media text data by keywords;

[0028] S2. Construct a semantic network of rockfall risk in scenic spots that integrates public sentiment;

[0029] S3. Based on the construction of a semantic network of rockfall risk in scenic spots that integrates public sentiment, analyze and identify scenic spots with high potential rockfall risks.

[0030] S1 includes obtaining social media text data related to scenic spots through the Internet.

[0031] S1 performs keyword filtering on social media text data, including deleting noise data and duplicate data, and extracting keywords. The keywords include public sentiment information, scenic spot names, and location information related to rockfall risks in social media text data, and constructs a scenic spot rockfall risk analysis database.

[0032] S2 includes sentiment information extraction and keyword analysis of social media text data based on the scenic spot rockfall risk analysis database, and integrates sentiment information and scenic spot information into the semantic network by calculating the node sentiment weight.

[0033] S2 constructs a semantic network of rockfall risk in scenic spots that integrates public emotions, including node construction that integrates public emotions, emotional node weight calculation, and scenic spot node score calculation.

[0034] S2 integrates the node construction of public emotions, including identifying emotional words and scenic spot words in social media text data through keyword screening, word frequency statistics and co-occurrence relationship analysis, and constructing emotional nodes and scenic spot nodes;

[0035] Emotional vocabulary includes positive emotion and negative emotion. The extracted emotional vocabulary is used as a sub-node in the semantic network. The sub-node types include positive emotion nodes and negative emotion nodes.

[0036] Scenic spot vocabulary includes scenic spot name and location information. The extracted scenic spot vocabulary is used as the parent node in the semantic network. The parent node types include positive sentiment nodes, neutral sentiment nodes and negative sentiment nodes.

[0037] S2 emotional node weight calculation includes calculating the number of positive emotional nodes S in all child nodes connected to the parent node p , the number of negative sentiment nodes S n , the weight of positive sentiment nodes W p and the weight of negative sentiment nodes W n .

[0038] The calculation of the S2 scenic spot node score includes calculating the weight ratio of negative emotion nodes to positive emotion nodes to obtain the score P of the parent node. When 0≤P<1, the parent node type is a positive emotion node, indicating that the public's positive emotion towards the scenic spot is greater than the negative emotion, and the risk of rockfall in the scenic spot is low risk, and the risk is lower when P is closer to 0; when P=1, the parent node type is a neutral emotion node, indicating that the risk of rockfall in the scenic spot is medium risk; when P>1, the parent node type is a negative emotion node, indicating that the public's negative emotion towards the scenic spot is greater than the positive emotion, and the risk of rockfall in the scenic spot is high risk, and the larger P is, the higher the risk.

[0039] In S2, the sentiment of the parent node is analyzed, a semantic network of rockfall risk in scenic spots that integrates public sentiment is constructed, the semantic network structure is optimized, and the scenic spots discussed by the public and the corresponding semantic relations are identified in combination with scenic spot detection to obtain the potential rockfall risk of scenic spots.

[0040] S3 includes calculating the value and weight of each child node in the semantic network, evaluating the association strength between the child node and the parent node, analyzing the sentiment type of the parent node, sorting the parent node according to the negative sentiment intensity, calculating the comprehensive risk index of the scenic area, and predicting and evaluating the potential rockfall risk in the scenic area.

[0041] In the semantic network, each node represents an independent information entity, including public sentiment and scenic spot information. In the semantic network, the sentiment node as a child node represents a piece of social media data containing sentiment vocabulary. Sentiment nodes are divided into positive sentiment nodes and negative sentiment nodes.

[0042] The embodiment of the present invention obtains Weibo data and Xiaohongshu data with "rockfall in scenic spots" as the keyword, constructs a semantic network of rockfall risk in scenic spots that integrates public sentiment, and analyzes the rockfall risk characteristics of specific scenic spots based on the results, thereby accurately predicting and evaluating the rockfall risk in scenic spots.

[0043] The semantic network diagram of rockfall risk in scenic spot A of the present invention is as follows Figure 2 As shown in Figure 2, the positive sentiment words extracted from the data obtained from Weibo and Xiaohongshu about scenic spot A are clear, safe, and relaxed, and the negative sentiment words are injured, killed, unfortunate, unexpected, collapsed, regretful, and thrilling. Through calculation, it can be concluded that S p 3, S n =7, Wp is 0.3, W n is 0.7, and the scenic spot node score P is 2.3, so it can be considered that the emotion of scenic spot A is negative and the risk of rockfall is high.

[0044] The semantic network diagram of rockfall risk in scenic area B of the present invention is as follows Figure 3 As shown in Figure 2, the positive sentiment words extracted from the data obtained from Weibo and Xiaohongshu about scenic spot B are safe, beautiful, comfortable, beautiful scenery, and happy, and the negative sentiment words are injured, scratched, fell, panic, and stepped on. Through calculation, it can be concluded that S p is 5, S n =5,W p =0.5, W n is 0.5, and the scenic spot node score P is 1, so it can be considered that the sentiment of scenic spot B is neutral and the rockfall risk is medium.

[0045] The semantic network diagram of rockfall risk in scenic area C of the present invention is as follows Figure 4 As shown in Figure 2, the positive sentiment words extracted from the data obtained from Weibo and Xiaohongshu about scenic spot C are cool, beautiful, absolutely beautiful, like, pretty, great value, exquisite, and the negative sentiment words are garbage, hard, and troublesome. Through calculation, it can be concluded that S p is 7, S n =3, W p is 0.7, W n is 0.3, and the scenic spot node score P is 0.43, so it can be considered that the emotion of scenic spot C is positive and the risk of rockfall is low.

[0046] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features may be replaced by equivalents, and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for investigating rockfall risks in scenic spots based on public observation, characterized in that: include: S1. Obtaining the scenic spot social media text data and filtering the social media text data by keywords; S2. Construct a semantic network of rockfall risk in scenic spots that integrates public sentiment; S3. Based on the construction of a semantic network of rockfall risk in scenic spots that integrates public sentiment, analyze and identify scenic spots with high potential rockfall risks.

2. A method for investigating rockfall risks in scenic areas based on public observation according to claim 1, characterized in that: S1 includes obtaining social media text data related to scenic spots through the Internet.

3. The method for investigating rockfall risks in scenic spots based on public observation according to claim 2 is characterized in that: S1 performs keyword filtering on social media text data, including deleting noise data and duplicate data, and extracting keywords. The keywords include public sentiment information, scenic spot names, and location information related to rockfall risks in social media text data, and constructs a scenic spot rockfall risk analysis database.

4. A method for investigating rockfall risks in scenic areas based on public observation according to claim 3, characterized in that: S2 includes sentiment information extraction and keyword analysis of social media text data based on the scenic spot rockfall risk analysis database, and integrates sentiment information and scenic spot information into the semantic network by calculating the node sentiment weight.

5. The method for investigating rockfall risks in scenic spots based on public observation according to claim 3 is characterized in that: S2 constructs a semantic network of rockfall risk in scenic spots that integrates public emotions, including node construction that integrates public emotions, emotional node weight calculation, and scenic spot node score calculation.

6. A method for investigating rockfall risks in scenic areas based on public observation according to claim 5, characterized in that: S2 integrates the node construction of public emotions, including identifying emotional words and scenic spot words in social media text data through keyword screening, word frequency statistics and co-occurrence relationship analysis, and constructing emotional nodes and scenic spot nodes; Emotional vocabulary includes positive emotion and negative emotion. The extracted emotional vocabulary is used as a sub-node in the semantic network. The sub-node types include positive emotion nodes and negative emotion nodes. Scenic spot vocabulary includes scenic spot name and location information. The extracted scenic spot vocabulary is used as the parent node in the semantic network. The parent node types include positive sentiment nodes, neutral sentiment nodes and negative sentiment nodes.

7. A method for investigating rockfall risks in scenic areas based on public observation according to claim 6, characterized in that: S2 emotional node weight calculation includes calculating the number of positive emotional nodes S in all child nodes connected to the parent node p , the number of negative sentiment nodes S n , the weight of positive sentiment nodes W p and the weight of negative sentiment nodes W n .

8. The method for investigating rockfall risks in scenic spots based on public observation according to claim 7, characterized in that: S2 Scenic spot node score calculation includes calculating the weight ratio of negative emotion nodes to positive emotion nodes to obtain the parent node score P. When 0≤P<1, the parent node type is a positive emotion node, indicating that the public's positive emotion towards the scenic spot is greater than the negative emotion, and the risk of rockfall in the scenic spot is low. The closer P is to 0, the lower the risk. When P = 1, the parent node type is a neutral sentiment node, indicating that the risk of rockfall in the scenic area is medium risk; When P>1, the parent node type is a negative sentiment node, indicating that the public's negative sentiment towards the scenic spot is greater than their positive sentiment, the risk of rockfall in the scenic spot is high, and the larger P is, the higher the risk.

9. The method for investigating rockfall risks in scenic spots based on public observation according to claim 8, characterized in that: In S2, the sentiment of the parent node is analyzed, a semantic network of rockfall risk in scenic spots that integrates public sentiment is constructed, the semantic network structure is optimized, and the scenic spots discussed by the public and the corresponding semantic relations are identified in combination with scenic spot detection to obtain the potential rockfall risk of scenic spots.

10. A method for investigating rockfall risks in scenic areas based on public observation according to claim 9, characterized in that: S3 includes calculating the value and weight of each child node in the semantic network, evaluating the association strength between the child node and the parent node, analyzing the sentiment type of the parent node, sorting the parent node according to the negative sentiment intensity, calculating the comprehensive risk index of the scenic area, and predicting and evaluating the potential rockfall risk in the scenic area.

Images