Innovation space identification method and device based on innovation subject and innovation activity data, and storage medium

Abstract

The invention belongs to the technical field of urban planning and construction, and particularly relates to an innovation space identification method for innovation subject and innovation activity data, which comprises the following steps of: gridding a map through an ArcGIS (Geographic Information System); acquiring patent data information including patent applicants, patent applicant addresses, patent application quantity and patent evaluation weights, and importing the patent data information into ArcGIS; obtaining data of an innovation subject and a carrier to construct an innovation energy level evaluation system and weights of the innovation subject and the carrier, and importing the data and the weights of the innovation subject and the carrier into an ArcGIS; carrying out breakpoint preprocessing; constructing a spatial weight matrix based on a polygon adjacency relation by using a Queen method of a pyal library; global Moran's I is calculated, the innovation space autocorrelation of the whole research area is evaluated, and the value and p-value of the global Moran's I are output; extracting hot spots, cold spots and abnormal values according to the four-quadrant classification of the local Moran's I, and filtering the abnormal values; and drawing an innovative space unit diagram based on the original data in the step 2 and the step 3, performing superposition, and taking an intersection.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban planning and construction, and specifically relates to an innovation space identification method for innovation subjects and innovation activity data. Background Art

[0002] At present, the conventional method for identifying innovation space is to identify and judge from a single patent perspective, which has the following defects: Although patent data covers many innovative activities in the technology field, it does not fully represent all innovative behaviors. For example, some companies may choose to protect their technology through trade secrets rather than patents, especially in the fields of software development, business methods, etc. In addition, non-patent forms of innovation such as academic research and open source projects may not be included in the analysis, resulting in an incomplete understanding of the innovation space.

[0003] The quality and value of patents vary widely. Some patents may simply be defensive patents filed to block competitors and do not represent real technological innovation. Failure to effectively distinguish between high-quality and low-quality patents when mapping the innovation space may lead to biased assessments of certain technology areas.

[0004] Innovation is not only the result of technological progress, but also affected by a variety of non-technical factors such as market demand, policy environment, cultural background, etc. However, these factors are difficult to directly reflect through patent data, so when using patent data to construct an innovation space map, these important external variables may be ignored. Summary of the invention

[0005] In view of this, the purpose of this invention is to take the information of the innovation subject into consideration in order to further improve the recognition accuracy of the innovation space and obtain more accurate information: On one hand, the present invention discloses an innovation space identification method based on innovation subject and innovation activity data, and the process is as follows: Step 1. Grid map through ArcGIS; Step 2. Obtain patent data information including patent applicant, address of patent applicant, number of patent applications, and patent evaluation weight and import them into ArcGIS; Step 3. Obtain the data of innovation subjects and carriers to build an innovation level evaluation system and the weights of innovation subjects and carriers, and import the data and weights of innovation subjects and carriers into ArcGIS; Step 4. Breakpoint preprocessing: classify the original data of step 2 and step 3 using the natural breakpoint method; Step 5. Use the Queen method of the pysal library to construct a spatial weight matrix based on polygon adjacency; Step 6. Calculate the global Moran's I, evaluate the spatial autocorrelation of innovation in the entire study area, and output the value and p-value of the global Moran's I; Step 7. According to the four-quadrant classification of local Moran's I, hot spots, cold spots and outliers are extracted, and outliers are filtered; Step 8. Draw the innovation space unit map based on the original data of steps 2 and 3, and superimpose them to obtain the intersection.

[0006] Among them, the patent evaluation weight a=b·number of patent applications+c·number of patent authorizations, where b and c are the first index and the second index respectively; Furthermore, the innovation subjects and carriers include: a total of 7 innovation subjects, including colleges and universities, research institutes, top 100 companies in R&D investment (divided into top 10 and top 100), specialized and new small giants, high-tech enterprises with a scale of over 100 million (divided into tens of billions, tens of billions, and hundreds of millions), listed high-tech enterprises, and unicorns (divided into unicorns, future unicorns, and seed unicorns); a total of 7 innovation carriers, including major scientific and technological innovation platforms, national enterprise technology centers, provincial new R&D institutions, science and technology banks or science and technology credit specialized institutions, registered private equity, venture capital fund managers, and science and technology business incubators at or above the provincial level.

[0007] In one aspect, the present invention discloses a computer device, the computer device comprising: a memory and at least one processor, the memory storing instructions; The at least one processor calls the instructions in the memory to enable the computer device to execute the innovation space identification method for innovation subjects and innovation activity data.

[0008] The present invention also discloses a computer-readable storage medium, on which instructions are stored, and when the instructions are executed by a processor, the innovation space identification method of the innovation subject and innovation activity data is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a specific implementation flow chart of the present invention; Figure 2 Draw an innovation space unit diagram based on step 2 in the embodiment; Figure 3 Draw an innovation space unit diagram based on step 3 in the embodiment; Figure 4 Draw an innovation space unit diagram based on step 7 in the embodiment; Figure 5 Draw an innovation space unit diagram based on step 8 in the embodiment; DETAILED DESCRIPTION

[0010] The content of the present invention will be further described below through specific implementation methods.

[0011] The present invention takes a certain region as an example to further illustrate the method of the present invention.

[0012] Step 1. Grid the map using ArcGIS; In order to make the grid scale more accurate in describing the layout of innovative space, the present invention refers to the existing practice and divides the area into 29755 500m×500m grids as the basic research unit. After experimental analysis and visualization test, the division of grid size meets the research needs.

[0013] Step 2. Obtain patent data information including patent applicant, address of patent applicant, number of patent applications, and patent evaluation weight and import them into ArcGIS; The specific computer instructions are as follows: import pandas as pd from geopy.geocoders import Nominatim import arcpy csv_file = 'patents.csv' df = pd.read_csv(csv_file) geolocator = Nominatim(user_agent="patent_geocoder") df['Latitude'] = None df['Longitude'] = None for index, row in df.iterrows(): try: location = geolocator.geocode(row['Applicant_Address']) if location: df.at[index, 'Latitude'] = location.latitude df.at[index, 'Longitude'] = location.longitude except Exception as e: print(f"Error geocoding address {row['Applicant_Address']}:{e}") output_gdb = r'C:\path\to\your\geodatabase.gdb' output_feature_class = 'Patent_Points' if arcpy.Exists(output_gdb + '\\' + output_feature_class): arcpy.Delete_management(output_gdb + '\\' + output_feature_class) arcpy.CreateFeatureclass_management(output_gdb, output_feature_class,"POINT", spatial_reference=arcpy.SpatialReference(4326)) fields = [ ('Applicant_Name', 'TEXT'), ('Applicant_Address', 'TEXT'), ('Application_Count', 'LONG'), ('Evaluation_Weight', 'DOUBLE') ] for field_name, field_type in fields: arcpy.AddField_management(output_gdb + '\\' + output_feature_class, field_name, field_type) with arcpy.da.InsertCursor(output_gdb + '\\' + output_feature_class,['SHAPE@XY'] + [field[0] for field in fields]) as cursor: for index, row in df.iterrows(): if not pd.isna(row['Latitude']) and not pd.isna(row['Longitude']): point = (row['Longitude'], row['Latitude']) values ​​= [point, row['Applicant_Name'], row['Applicant_Address'], row['Application_Count'], row['Evaluation_Weight']] cursor.insertRow(values) print("Data has been successfully imported into ArcGIS.") Step 3. Obtain the data of innovation subjects and carriers to build an innovation level evaluation system and the weights of innovation subjects and carriers, and import the data and weights of innovation subjects and carriers into ArcGIS.

[0014] The innovation level evaluation system used in the embodiment is as follows:

[0015] The specific computer instructions are as follows: import pandas as pd from geopy.geocoders import Nominatim import arcpy csv_file = 'innovation_data.csv' df = pd.read_csv(csv_file) def calculate_weight(row): return (row['Application_Count'] ) + (row['Evaluation_Weight'] ) df['Weight'] = df.apply(calculate_weight, axis=1) geolocator = Nominatim(user_agent="innovation_geocoder") df['Latitude'] = None df['Longitude'] = None for index, row in df.iterrows(): try: location = geolocator.geocode(row['Address']) if location: df.at[index, 'Latitude'] = location.latitude df.at[index, 'Longitude'] = location.longitude except Exception as e: print(f"Error geocoding address {row['Address']}: {e}") output_gdb = r'C:\path\to\your\geodatabase.gdb' output_feature_class = 'Innovation_Points' if arcpy.Exists(output_gdb + '\\' + output_feature_class): arcpy.Delete_management(output_gdb + '\\' + output_feature_class) arcpy.CreateFeatureclass_management(output_gdb, output_feature_class,"POINT", spatial_reference=arcpy.SpatialReference(4326)) fields = [ ('Subject_Name', 'TEXT'), ('Carrier_Name', 'TEXT'), ('Address', 'TEXT'), ('Application_Count', 'LONG'), ('Evaluation_Weight', 'DOUBLE'), ('Weight', 'DOUBLE')] for field_name, field_type in fields: arcpy.AddField_management(output_gdb + '\\' + output_feature_class, field_name, field_type) with arcpy.da.InsertCursor(output_gdb + '\\' + output_feature_class,['SHAPE@XY'] + [field[0] for field in fields]) as cursor: for index, row in df.iterrows(): if not pd.isna(row['Latitude']) and not pd.isna(row['Longitude']): point = (row['Longitude'], row['Latitude']) values ​​= [point, row['Subject_Name'], row['Carrier_Name'], row['Address'], row['Application_Count'], row['Evaluation_Weight'],row['Weight']] cursor.insertRow(values) print("Data has been successfully imported into ArcGIS.") Step 4. Breakpoint preprocessing: classify the original data of step 2 and step 3 using the natural breakpoint method; Step 5. Use the Queen method of the pysal library to construct a spatial weight matrix based on polygon adjacency; Step 6. Calculate the global Moran's I, evaluate the spatial autocorrelation of innovation in the entire study area, and output the value and p-value of the global Moran's I; Step 7. According to the four-quadrant classification of local Moran's I, hot spots, cold spots and outliers are extracted, and outliers are filtered; Step 8. Draw the innovation space unit map based on the original data of steps 2 and 3, and superimpose them to obtain the intersection.

[0016] The specific computer instructions are as follows: import pandas as pd import geopandas as gpd from jenkspy import jenks_breaks import pysal.lib.weights as weights from pysal.explore.esda import Moran, Moran_Local import matplotlib.pyplot as plt data = gpd.read_file('innovation_data.csv') breaks = jenks_breaks(data['innovation_rate'], nb_class=5) data['innovation_rate_class'] = pd.cut(data['innovation_rate'], bins=breaks, labels=False, include_lowest=True) w = weights.Queen.from_dataframe(data) moran_global = Moran(data['innovation_rate'], w) print(f"Global Moran's I: {moran_global.I}, p-value: {moran_global.p_sim}") moran_local = Moran_Local(data['innovation_rate'], w) data['moran_local'] = moran_local.Is hotspots = data[moran_local.q == 1] # High-High coldspots = data[moran_local.q == 3] # Low-Low outliers_high_low = data[moran_local.q == 2] # High-Low outliers_low_high = data[moran_local.q == 4] # Low-High filtered_data = data[~data.index.isin(outliers_high_low.index) & ~data.index.isin(outliers_low_high.index)] fig, ax = plt.subplots(1, 1, figsize=(15, 10)) data.plot(ax=ax, column='innovation_rate_class', cmap='viridis',edgecolor='black', legend=True) hotspots.plot(ax=ax, color='red', alpha=0.5, label='Hot Spots') coldspots.plot(ax=ax, color='blue', alpha=0.5, label='Cold Spots') filtered_data.plot(ax=ax, color='green', alpha=0.5, label='FilteredData') ax.legend() ax.set_title('Innovation Rate Distribution and Hot / Cold Spots') ax.set_xlabel('Longitude') ax.set_ylabel('Latitude') plt.show() intersection = gpd.sjoin(filtered_data, data[['geometry', 'innovation_rate_class']], op='intersects') fig, ax = plt.subplots(1, 1, figsize=(15, 10)) data.plot(ax=ax, column='innovation_rate_class', cmap='viridis', edgecolor='black', legend=True) intersection.plot(ax=ax, color='purple', alpha=0.5, label='Intersection') ax.legend() ax.set_title('Innovation Rate Distribution and Intersection') ax.set_xlabel('Longitude') ax.set_ylabel('Latitude') plt.show() like Figure 2 The following is the innovation space unit diagram drawn in step 2: Figure 3 The following is the innovation space unit diagram drawn in step 3: Figure 4 Shown is the innovation space unit diagram obtained by executing steps 4-8.

[0017] The various component embodiments of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all of the functions of some modules in the control device according to an embodiment of the present invention. The present invention can also be implemented as a device program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

Claims

1. A method for identifying innovation space based on innovation subject and innovation activity data, characterized by: The steps include: Step 1. Grid the map using ArcGIS; Step 2. Obtain patent data information including patent applicant, address of patent applicant, number of patent applications, and patent evaluation weight and import them into ArcGIS; Step 3. Obtain the data of innovation subjects and carriers to build an innovation level evaluation system and the weights of innovation subjects and carriers, and import the data and weights of innovation subjects and carriers into ArcGIS; Step 4. Breakpoint preprocessing: classify the original data of step 2 and step 3 using the natural breakpoint method; Step 5. Use the Queen method of the pysal library to construct a spatial weight matrix based on polygon adjacency; Step 6. Calculate the global Moran's I, evaluate the spatial autocorrelation of innovation in the entire study area, and output the value and p-value of the global Moran's I; Step 7. According to the four-quadrant classification of local Moran's I, hot spots, cold spots and outliers are extracted, and outliers are filtered; Step 8. Draw the innovation space unit map based on the original data of steps 2 and 3, and superimpose them to obtain the intersection.

2. The innovation space identification method based on innovation subject and innovation activity data according to claim 1 is characterized by: The patent evaluation weight a=b· Number of applications for benefits + c · Number of patents granted, where b and c are the first index and the second index respectively.

3. The innovation space identification method based on innovation subject and innovation activity data according to claim 1 is characterized by: The innovation subjects and carriers mentioned include: a total of 7 innovation subjects, including universities, research institutes, top 100 companies in R&D investment, specialized and innovative small giants, high-tech enterprises with a scale of over 100 million yuan, listed high-tech enterprises, and unicorn enterprises; a total of 7 innovation carriers, including major scientific and technological innovation platforms, national enterprise technology centers, provincial new R&D institutions, science and technology banks or science and technology credit specialized institutions, registered private equity, venture capital fund managers, and science and technology business incubators at or above the provincial level.

4. A computer device, characterized in that: The computer device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the computer device to execute the innovation space identification method for innovation subject and innovation activity data as described in any one of claim 1.

5. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the innovation space identification method of the innovation subject and innovation activity data as described in any one of claim 1 is implemented.

Images