A method and system for extracting spatiotemporal and semantic information based on ontology
Through the ontology-based space-time and semantic information extraction method, using technologies such as place directory matching and regular expressions, we automatically process geological disaster reports to form structured information, solving the problem of low data processing efficiency of geological disaster reports and achieving efficient information Extract and visual rendering.
Patent Information
- Application Number
- CN202111080748.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-09-15
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2041-09-15
AI Technical Summary
Geological disaster reporting has large amounts of data, high noise, fast update speed, and strong language arbitraryness, making it difficult for computers to process directly. The existing technology relies on a large amount of manual analysis and is inefficient and cannot meet the needs of quickly obtaining key information.
The spatial and semantic information extraction method based on the ontology is used to extract the spatial and semantic information in the geological disaster report through place directory matching, regular expressions and ontology matching, and form a structured quadruple and visual presentation.
It realizes automated information extraction and visualization, improves the efficiency and accuracy of geological disaster reporting information acquisition, reduces manual intervention, and meets the needs of rapid understanding and analysis.
Smart Images

Figure CN113919296B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of data mining and natural language processing, and in particular to an ontology-based spatiotemporal and semantic information extraction method and system. Background Art
[0002] In the past few years, in addition to maps, satellite images and global positioning systems, text has become an important data resource. Extracting information from geological disaster texts has become a research hotspot, and its value has gradually become prominent. Obtaining key data from a large amount of geological disaster texts and fully mining the knowledge and information implicit in unstructured data can enable people to quickly obtain key information from geological disaster reports. The general definition of information extraction refers to processing the information included in the text to obtain structured information, which can be transformed into a table-like organizational form. The purpose of information extraction research is to discover the useful attributes of specific types of these tags and their interrelationships, provide people with more reliable information acquisition tools, and use these tools to make the information obtained more and more accurate.
[0003] Geological disaster reports contain a large amount of relevant information about disaster events. Extracting and utilizing this information can effectively understand the disaster information, communication status, casualties and losses of disaster events, and provide a timely and reliable source of information for making emergency decisions in the first place. However, due to the huge amount of data, high noise, fast update speed, and the randomness and non-structured characteristics of geological disaster reports, it is difficult for computers to process them directly.
[0004] At present, many text analysis tasks still require a lot of manual work to complete. Due to the frequent occurrence of geological disasters, the number of disaster reports will increase, and the efficiency of processing tasks will decrease. A geological disaster report contains a lot of content, and reading and recording is not an easy task. The speed of manual information extraction can no longer meet the current needs of people for analysis, understanding and application. Summary of the invention
[0005] One of the main problems solved by the present invention is the problem of the ever-increasing workload of text analysis.
[0006] According to one aspect of the present invention, the present invention provides a method for extracting spatiotemporal and semantic information based on ontology, comprising:
[0007] Convert the geological disaster report text into a preset format to obtain a converted text;
[0008] Removing the graphic content in the converted text, and dividing the converted text content into a plurality of independent input units in the form of paragraphs, to obtain an input text;
[0009] Performing word segmentation on the input text to obtain a word segmentation text;
[0010] Using a place name directory matching method, the unit words in the segmented text are matched one by one with the place names in the place name directory to obtain spatial information;
[0011] Extracting semantic information from the input text using regular expressions and ontology matching;
[0012] Combining the obtained spatial information and the obtained semantic information in units of sentences to form a plurality of quadruples;
[0013] Filter out a valid quadruple from the plurality of quadruple groups;
[0014] The valid quadruple is visualized.
[0015] Furthermore, the word segmentation of the input text by using the place name directory matching method includes:
[0016] Subdividing the plurality of independent input units into independent sentences, inputting the independent sentences, and segmenting the independent sentences using the jieba library to obtain the segmented texts;
[0017] The unit words in the obtained segmented text are matched one by one with the place names in the place name directory to obtain the geographical location where the geological disaster occurred.
[0018] Furthermore, the word segmentation of the input text by using the place name directory matching method also includes:
[0019] When the unit word fails to match the place name, a qualifier is added after the unit word and then the match is repeated.
[0020] Furthermore, the word segmentation of the input text by using the place name directory matching method also includes:
[0021] When the unit word fails to match the place name, the unit word is matched with words in a synonym word library.
[0022] Furthermore, the method of extracting semantic information from the input text by using regular expressions and ontology matching includes:
[0023] Using regular expressions, each unit word in the segmented text is matched with a word in the causal word library. If the match is successful, all the contents between each unit word and the end of the sentence are recorded as the impact information caused by the disaster;
[0024] Using the ontology matching method, if a unit word in the segmented text successfully matches the ontology, the two words in front of the unit word are matched with the words in the causal word library. If the match is successful, the unit word is stored as the impact information caused by the disaster. If the match is unsuccessful, the unit word is stored as the geological disaster type information.
[0025] Furthermore, the effective quadruple contains at least three elements: geological disaster type information, time information and space information.
[0026] Furthermore, the spatiotemporal and semantic information extraction method further includes obtaining time information using a regular expression:
[0027] Using regular expressions, the first time information extracted from each paragraph of the independent input unit is used as the paragraph time of each paragraph. If the subsequent sentences of each paragraph lack time information, the paragraph time is used as the time information of the geological disaster event in the subsequent sentences.
[0028] According to another aspect of the present invention, a spatiotemporal and semantic information extraction system based on ontology is also disclosed, which includes a storage module, a processing module and an execution module. The storage module stores a computer program, the processing module is used to process geological disaster report text data, and the execution module is used to call the computer program to execute any of the ontology-based spatiotemporal and semantic information extraction methods described above.
[0029] The present invention extracts unstructured information from geological disaster reports through methods such as named entity recognition, ontology matching, and place name matching to form structured information; finally, the final results extracted from the geological disaster reports are presented in a reasonable visualization manner, which can help people better understand and explore the complex text information contained in big data, and express it in the form of simple visual symbols to achieve the purpose of human-computer interaction. The method has high practicality and reference value in the extraction and visualization of geological disaster report texts. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.
[0031] Figure 1 It is a flow chart of a method for extracting spatiotemporal and semantic information based on ontology in an embodiment of the present invention;
[0032] Figure 2 A visualization graph generated using the extracted quadruple as data in an embodiment of the present invention; DETAILED DESCRIPTION
[0033] Various exemplary embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention.
[0034] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.
[0035] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.
[0036] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.
[0037] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.
[0038] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.
[0039] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.
[0040] Embodiment 1, as Figure 1 As shown, Figure 1 The flowchart of a method for extracting spatiotemporal and semantic information based on ontology according to an embodiment of the present invention specifically includes the following steps:
[0041] S1. Obtaining a geological disaster report, and preprocessing the geological disaster report to obtain a preprocessed geological disaster report text;
[0042] S2, extracting spatial information from the pre-processed geological disaster report text by using a place name matching method;
[0043] S3, extracting semantic information from the pre-processed geological disaster report text using regular expressions and ontology matching;
[0044] S4, combining the extracted spatial information and semantic information in units of sentences to form a number of quadruple groups (including geological disaster type information, time information, location information, and disaster impact information), each of which can represent a complete geological disaster event;
[0045] S5. Filter out valid quadruplets and visualize them.
[0046] Specifically, in step S1, the preprocessing of the geological disaster report includes: format conversion of the geological disaster report text, such as converting the geological disaster report PDF format into the word format, and removing the charts and other contents in the word file; considering that a paragraph in a geological disaster report usually describes the same event, but the description angles are different, but there is a strong correlation between the sentences, so the word file content is divided into paragraphs, and each paragraph is used as an independent input unit.
[0047] Specifically, in step S2, the place name directory matching includes:
[0048] S21, further subdividing the input unit based on the paragraph, taking the sentence as the input unit, and using the ji eba library to segment the input sentence text;
[0049] S22, matching the segmented words obtained above with the place names in the place name directory one by one to obtain the geographical location information of the geological disaster;
[0050] S23. Considering the abbreviation phenomenon in Chinese, for example, Shanxi Province is abbreviated as Shanxi, while the place names in the place name directory are relatively complete place names, and in this case, it is impossible to successfully match. In order to deal with the abbreviation problem, when the segmentation word and the place name are not matched successfully, "province", "city" or "county" is added after the segmentation word and then matched.
[0051] S24, considering that Chinese has many different names for the same place, and the place names in the place name directory are all more official place names, when the method of step S23 above fails to match, the words in the word segmentation text are matched with the words in the synonym word library. The word library is divided into two categories: original words and derived words. The original words are more formal place names, and the derived words are other names for the place name. There is a corresponding relationship between the original words and the derived words, and one original word can correspond to multiple derived words. For example, Hubei Province belongs to the original word in the synonym word library, and it corresponds to the derived words such as "E" and "Jingchu". When the word segmentation matches the derived word, the original word corresponding to the derived word will be stored.
[0052] Specifically, step S3 also includes: extracting time information and disaster impact information from the unsegmented text through regular expressions, the extractable time information format is shown in Table 1, and the extractable disaster impact information format is shown in Table 2. Matching the segmented text with the words in the causal word library, the words in the library are all words that represent causal relationships, such as "lead to", "cause", "trigger", etc. If the match is successful, the sentence content from the beginning of the word to the end of this part will be recorded as the impact information caused by the disaster; extracting specific disaster type information from the segmented input text in an ontology-based matching manner, if a word in the segmented text successfully matches the ontology, the two words before the word are matched with the words in the causal word library, if the match is successful, it means that the disaster is caused by other disasters, and the word is stored as the impact information, if the first two words do not match successfully, the word is stored as the geological disaster type information.
[0053] Table 1 Regular expressions for time information extraction
[0054]
[0055] Table 2 Regular expressions for extracting the impact of disasters
[0056]
[0057] Specifically, step S4 includes: integrating the information of geological disaster, time, location and disaster impact extracted in the above steps into four-tuples in units of sentences. Among them, since some geological disaster texts will give time information at the beginning of a paragraph, the subsequent sentences will not be accompanied by time information, but the time given at the beginning of the paragraph is assumed to be the time when the geological disaster occurs later. Therefore, the first time information extracted from each paragraph is used as the paragraph time of the paragraph. If the subsequent sentences of the paragraph lack time information, the paragraph time is used as the time when the geological disaster event in the sentence occurs.
[0058] Specifically, step S5 also includes: selecting quadruple groups containing at least three elements, namely geological disaster, time and location information, from the quadruple groups corresponding to the sentences, and then presenting these quadruple groups in the form of a knowledge graph, such as Figure 2 shown.
[0059] The beneficial effects of the present invention are as follows: the present invention provides an ontology-based spatiotemporal and semantic information extraction method for automatically identifying the semantic and spatiotemporal information of geological disaster reports. It solves the growing demand for text analysis workload and has high practicality and reference value for information extraction and information visualization of geological disaster report texts.
[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
[0061] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
Claims
1. A method for extracting spatiotemporal and semantic information based on ontology, characterized in that: include: Convert the geological disaster report text into a preset format to obtain a converted text; Removing the graphic content in the converted text, and dividing the converted text content into a plurality of independent input units in the form of paragraphs, to obtain an input text; Performing word segmentation on the input text to obtain a word segmentation text; Using a place name directory matching method, the unit words in the segmented text are matched one by one with the place names in the place name directory to obtain spatial information; Use regular expressions and ontology matching to extract semantic information from the input text; specifically: Using regular expressions, each unit word in the segmented text is matched with a word in the causal word library. If the match is successful, all the contents between each unit word and the end of the sentence are recorded as the impact information caused by the disaster; By using the ontology matching method, if a unit word in the segmented text successfully matches the ontology, the two words in front of the unit word are matched with the words in the causal word library. If the match is successful, the unit word is stored as the impact information caused by the disaster. If the match is unsuccessful, the unit word is stored as the geological disaster type information. Combining the obtained spatial information and the obtained semantic information in units of sentences to form a plurality of quadruples; Filter out a valid quadruple from the plurality of quadruple groups; The valid quadruple is visualized.
2. The ontology-based spatiotemporal and semantic information extraction method according to claim 1, characterized in that: The method of using the place name directory matching to segment the input text includes: Subdividing the plurality of independent input units into independent sentences, inputting the independent sentences, and segmenting the independent sentences using the jieba library to obtain the segmented text; The unit words in the obtained segmented text are matched one by one with the place names in the place name directory to obtain the geographical location where the geological disaster occurred.
3. The ontology-based spatiotemporal and semantic information extraction method according to claim 2, characterized in that: The method of using the place name directory matching method to segment the input text further includes: When the unit word fails to match the place name, a qualifier is added after the unit word and then the match is repeated.
4. The ontology-based spatiotemporal and semantic information extraction method according to any one of claims 2-3, characterized in that: The method of using the place name directory matching method to segment the input text further includes: When the unit word fails to match the place name, the unit word is matched with words in a synonym word library.
5. The ontology-based spatiotemporal and semantic information extraction method according to claim 1, characterized in that: The effective quadruple contains at least three elements: geological disaster type information, time information and space information.
6. The ontology-based spatiotemporal and semantic information extraction method according to claim 5, characterized in that: The method for extracting spatiotemporal and semantic information also includes obtaining time information using a regular expression: Using regular expressions, the first time information extracted from each paragraph of the independent input unit is used as the paragraph time of each paragraph. If the subsequent sentences of each paragraph lack time information, the paragraph time is used as the time information of the geological disaster event in the subsequent sentences.
7. A spatiotemporal and semantic information extraction system based on ontology, characterized in that: The spatiotemporal and semantic information extraction system includes a storage module, a processing module and an execution module. The storage module stores a computer program. The processing module is used to process the text data of geological disaster reports. The execution module is used to call the computer program to execute the ontology-based spatiotemporal and semantic information extraction method as described in claim 1.
