Mine hidden disaster-causing factor automatic identification method based on text mining

Abstract

The invention discloses an automatic mine hidden disaster-causing factor identification method based on text mining. The method comprises the following steps: matching hidden disaster-causing factor information based on keywords; dynamic generation of report options is realized; calculating and realizing total word frequency statistics; calculating sub-word frequency statistics; browsing the report at the front end; and screening and displaying the front-end report. A plurality of mine reports, rules and regulations are in butt joint, text information in the reports is crawled and obtained in real time, a text mining device is constructed, the reports under an integrated folder are all transmitted to a front end for display, the reports are autonomously selected for fuzzy search of mine hidden disaster-causing factors, and a report word frequency statistical graph is autonomously generated.

Description

Technical Field

[0001] The present invention relates to a technology for identifying and preventing coal mine safety accidents, and in particular to a method for automatically identifying hidden disaster-causing factors in mines based on text mining. Background Art

[0002] Coal resources have long been the most widely used and important energy source. Coal resources have long provided an important driving force for the sustained growth of the global economy. There are many safety issues and hidden dangers involved in the process of coal mining. Therefore, safe coal mining is still an important foundation for energy security and economic stability.

[0003] At present, although there have been many studies on risk identification, evaluation, risk response and control technology in recent years, the ability to obtain text data from coal mine production was weak, the means of text data analysis was single, the computing power of computer equipment was limited, and the research progress of text resources was slow. At present, most coal mining companies still analyze the hidden disaster factors of accidents based on the knowledge reserves and subjective judgments of on-site personnel and the comparison of indicator thresholds calculated in advance. Prevention and control measures are mostly based on historical experience and lessons for conventional safety decisions, such as:

[0004] You Mengjie. Research on coal mine safety risk identification and evaluation based on text mining[D]. China University of Mining and Technology, 2022;

[0005] At present, the investigation and management of hidden disaster-causing factors of coal mine accidents mainly focuses on the process control level, ignoring the shortcomings in the management of coal mine accident hazards through research and utilization of hidden danger text data information, so that the massive coal mine accident hidden danger investigation text data accumulated over a long period of time has not been truly utilized, such as:

[0006] Zhao Ke. Research on application of coal mine accident hazard investigation information system based on text mining[D]. China University of Mining and Technology, 2020.DOI:10.27623 / d.cnki.gzkyu.2020.001560.

[0007] The amount of data on coal mine water damage is huge. In the process of compliance review of reports and materials, and demonstration of water damage prevention and control plans, it is a heavy task to quickly read and sort out the massive data. In this context, text mining technology in current information technology can be used to extract keywords from massive water damage data, explore hidden disaster factors in mines, realize text analysis in specific scenarios, and assist process management.

[0008] In view of this, the present invention is proposed. Summary of the invention

[0009] The purpose of the present invention is to provide a method for automatically identifying hidden disaster-causing factors in mines based on text mining to solve the above-mentioned technical problems existing in the prior art.

[0010] The objective of the present invention is achieved through the following technical solutions:

[0011] The method for automatically identifying hidden disaster-causing factors in mines based on text mining of the present invention comprises:

[0012] Keyword-based matching of hidden disaster-causing factors information;

[0013] Implement dynamic generation of report options;

[0014] Calculation and implementation of total word frequency statistics;

[0015] Calculate subword frequency statistics;

[0016] Front-end browsing report;

[0017] Filtered display of front-end reports.

[0018] Compared with the prior art, the method for automatic identification of hidden disaster-causing factors in mines based on text mining provided by the present invention connects to a number of mine reports and rules and regulations, crawls the text information therein in real time, constructs a text mining device, transmits all reports in the integrated folder to the front-end display, independently selects reports for fuzzy search of hidden disaster-causing factors in mines, and independently generates report word frequency statistics. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of fuzzy search of hidden disaster-causing factors based on keywords in an embodiment of the present invention;

[0020] Figure 2 A schematic diagram of a process for dynamically generating report options according to an embodiment of the present invention;

[0021] Figure 3 A schematic diagram of a process for calculating total word frequency statistics according to an embodiment of the present invention;

[0022] Figure 4 A schematic diagram of a process for calculating subword frequency statistics according to an embodiment of the present invention;

[0023] Figure 5 A schematic diagram of the process of front-end browsing reports according to an embodiment of the present invention;

[0024] Figure 6 A schematic diagram of the process of screening and displaying the front-end report according to an embodiment of the present invention;

[0025] Figure 7 The present invention is a schematic diagram of the use of a method for automatically identifying hidden disaster-causing factors in mines based on text mining according to an embodiment of the present invention.

[0026] Figure 8 The figure is a schematic diagram of the construction of the AGWS according to the embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention; it is obvious that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, which does not constitute a limitation of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of the present invention.

[0028] First, the terms that may be used in this article are explained as follows:

[0029] The term “and / or” means that either or both of them can be realized at the same time. For example, X and / or Y means both “X” or “Y” and “X and Y”.

[0030] The terms "include", "comprises", "contains", "has" or other descriptions with similar semantics should be interpreted as non-exclusive inclusion. For example, including certain technical feature elements (such as raw materials, components, ingredients, carriers, dosage forms, materials, dimensions, parts, components, mechanisms, devices, steps, procedures, methods, reaction conditions, processing conditions, parameters, algorithms, signals, data, products or products, etc.) should be interpreted as including not only certain technical feature elements explicitly listed, but also other technical feature elements known in the art that are not explicitly listed.

[0031] The contents not described in detail in the examples of the present invention belong to the prior art known to professionals in the field. If no specific conditions are specified in the examples of the present invention, the conditions are carried out according to the conventional conditions in the field or the conditions recommended by the manufacturer. If the manufacturer is not specified in the reagents or instruments used in the examples of the present invention, they are all conventional products that can be purchased commercially.

[0032] The method for automatically identifying hidden disaster-causing factors in mines based on text mining of the present invention comprises:

[0033] Keyword-based matching of hidden disaster-causing factors information;

[0034] Implement dynamic generation of report options;

[0035] Calculation and implementation of total word frequency statistics;

[0036] Calculate subword frequency statistics;

[0037] Front-end browsing report;

[0038] Filtered display of front-end reports.

[0039] The keyword-based hidden disaster factor information matching includes:

[0040] Definition: Establish the Geological Word Set (GWS) and the General Geological Word Set (AGWS);

[0041] Step 1: Obtain the report name set (RNS) to be queried and the hidden disaster-causing key factors (HKEY) to be queried;

[0042] Step 2: According to the report name set (RNS), obtain the report set (RS) to be queried from the entire report set (REPORT);

[0043] Step 3: Get the i-th report RS in the report set (RS) i Clean complex symbols in the report, such as special geological age symbols, unit water inflow units, etc., and replace RS i Split the report text according to its sentence structure and return the geological sentence set (GSS) of the report.

[0044] Step 4: Get the jth statement GSS in the geological statement set (GSS) j , identify and extract GSS j All words in the Geological Word Set (GWS) are merged into the Geological Word Set (GWS);

[0045] Step 5: Get the kth set GWS in GWS k , incorporated into the General Set of Geological Words (AGWS);

[0046] Step 6: Repeat steps 4-5;

[0047] Step 7: Process the hidden disaster-causing key factors (HKEY) as in steps 4-6 to obtain a disaster-causing query word set (HKEYS);

[0048] Step 8: Set the parameters of the algorithm. k1, k2, and b are coordination factors, which are set to 2, 1, and 0.75 respectively.

[0049] Step 9: Calculate the similarity between the total set of geological words (AGWS) and the hidden disaster-causing key factors (HKEY): n represents the total number of words in HKEYS, W i Indicates HKEYS i The weight owned, R(HKEYS i , GSS j ) represents the relevance score between the i-th word in HKEYS and a sentence in the geological sentence set (GSS), where N represents the total number of statements in the geological statement set (GSS), dfi Indicates that HKEYS is included i The total number of sentences in the geological report, of which in f i Indicates HKEYS i The total number of times a sentence appears in a geological report. dl represents GSS j The length of the statement, avg_dl represents the average length of the Geological Statement Set (GSS) statements;

[0050] Step 10: Define the similarity set (SS);

[0051] Step 11: Add the similarity score to the similarity set;

[0052] Step 12: Repeat step 11;

[0053] Step 13: Sort the similarity set (SS) to obtain the first five corresponding geological statements in the geological statement set (GSS).

[0054] The dynamic generation of report options includes:

[0055] Step 1: Extract the set of all report names (RNS) from the set of all reports (REPORT);

[0056] Step 2: Load all report name sets (RNS) into the hidden disaster factor mining device.

[0057] The calculation and implementation of the total word frequency statistics include:

[0058] Step 1: Get the report name set (RNS) to be counted, establish the result word set (WS), establish the result word frequency set (WF), and establish the mapping relationship f of the set: WS->WF;

[0059] Step 2: Obtain the report set (RS) to be counted from the entire report set (REPORT) according to the report name set (RNS);

[0060] Step 3: Get the i-th report RS in the report set (RS) i Extract all text content;

[0061] Step 4: Add the text content extracted in step 3 to the total text collection (ATEXT);

[0062] Step 5: loop steps 3-4 until all text contents in the report set (RS) are added to the total text set (ATEXT);

[0063] Step 6: Introduce the stop word set (SWORDS) and the target word set (TWORDS);

[0064] Step 7: Store each line in the entire text set (ATEXT) as a line set (ATEXTLINE);

[0065] Step 8: Extract ATEXTLINE i All words in SWORDS, if the word does not belong to SWORDS j , and the word belongs to TWORDS j , the word is added to WS k The corresponding WF under the mapping relationship f k Add 1 to the value;

[0066] Step 9: Repeat step 8;

[0067] Step 10: Loop steps 8-9 until the row set (ATEXTLINE) is exhausted;

[0068] Step 11: Referring to the top 20 largest words in the result word frequency set (WF), according to the mapping relationship f, extract the corresponding result word set (WS) subset and result word frequency set (WF) subset.

[0069] The calculation of subword frequency statistics includes:

[0070] Step 1: Get the report name set (RNS) to be counted, establish the result word set (WS), establish the result word frequency set (WF), and establish the mapping relationship f of the set: WS->WF;

[0071] Step 2: Obtain the report set (RS) to be counted from the entire report set (REPORT) according to the report name set (RNS);

[0072] Step 3: Get the i-th report RS in the report set (RS) i Extract all text content;

[0073] Step 4: Introduce the stop word set (SWORDS) and the target word set (TWORDS);

[0074] Step 5: RS i Each row is stored as a row set (TEXTLINE);

[0075] Step 6: Extract TEXTLINE i All words in SWORDS, if the word does not belong to SWORDS j , belongs to TWORDS j , the word is added to WS k , the corresponding WF under f kAdd 1 to the value;

[0076] Step 7: Repeat step 6;

[0077] Step 8: Repeat steps 3-7 until the report set (RS) is exhausted;

[0078] Step 9: Referring to the top 20 with the largest number in WF, according to the mapping relationship f, extract the corresponding result word set (WS) subset and result word frequency set (WF) subset.

[0079] The front-end browsing report includes:

[0080] Step 1: Get the report name (RN) to be browsed and define the report path set (RPS);

[0081] Step 2: According to RN, find the same vector in the entire report name set (RNS);

[0082] Step 3: According to f:RNS i ->RPS i , find the path to which the corresponding report name is mapped;

[0083] Step 4: Install according to RPS i Display the report content.

[0084] The screening display of the front-end report includes:

[0085] Step 1: Create a reporting mine set (RM) and a reporting category set (RV);

[0086] Step 2: Create a set of all reports (REPORT);

[0087] Step 3: According to the mapping relationship f:RM i ,RV j ->REPORT k , get the report subset (RSUB);

[0088] Step 4: Create a report name subset (RNSUB) based on RSUB;

[0089] Step 5: The device displays all file names after the filter category using Report Name Subset (RNSUB).

[0090] In summary, the method for automatically identifying hidden disaster-causing factors in mines based on text mining in the embodiment of the present invention connects to several mine reports and rules and regulations, crawls the text information therein in real time, and outputs multiple results in the order of similarity scores based on the Bm25 correlation analysis improved by tf-idf;

[0091] Convert the text into a corpus, compare the similarity between the queried disaster-causing factors and the corpus, and output the most suitable results. Filter out hidden disaster-causing factors, establish a hidden disaster-causing set, filter out those that are not in the set, and improve the accuracy;

[0092] Output the fuzzy matching results of hidden disaster-causing factors of each report, generate total word frequency statistics (total word cloud diagram), sub-word frequency statistics (sub-word cloud diagram), and analyze the overall and individual reports;

[0093] A text mining device was built, and all reports in the integrated folder were transmitted to the front-end display, reports were independently selected for fuzzy search of hidden disaster-causing factors in mines, and report word frequency statistics were independently generated.

[0094] In order to more clearly demonstrate the technical solution and technical effects provided by the present invention, the embodiments of the present invention are described in detail with specific embodiments below.

[0095] Example 1

[0096] 1. Implementation of main technical methods

[0097] 1. If Figure 1 As shown in the figure, the hidden disaster factor information matching based on keywords

[0098] The specific steps are as follows:

[0099] Definition: Establish a geological word set (GWS), establish an overall geological word set (AGWS).

[0100] Step 1: Get the report name set (RNS) to be queried, which represents a simple name text, i.e., {"A Mine Hidden Disaster Factor Survey and Control Report", "B Mine Hidden Disaster Factor Survey and Control Report", "C Mine Hidden Disaster Factor Survey and Control Report High"}, and the hidden disaster key factor (HKEY) to be queried, i.e., {"Water accumulation in the goaf"}.

[0101] Table 1

[0102] The report name set RNS to be queried Report on the Survey and Control of Hidden Disaster-causing Factors in A Mine Report on the Survey and Control of Hidden Disaster-causing Factors in B Mine Report on the investigation and control of hidden disaster-causing factors in C mine

[0103] HKEY: Water in the Goaf

[0104] Table 2

[0105] REPORTAll reports are collected <![CDATA[Report on the Hydrogeological Type Division of Mine A. pdf > <![CDATA[Mine Geological Environment Protection and Land Reclamation Plan. pdf > <![CDATA[3D Geological Exploration Report of the East Wing of A Coal Mine井田. pdf > <![CDATA[Transient Electromagnetic Exploration Report of Working Face 2102 in A Coal Mine. pdf > <![CDATA[Research Report on the Detection of the "Three Zones" Heights in the 3101 Working Face of A Coal Mine. pdf > <![CDATA[Summary Report on the Detection of the Wide Falling Fissure Zone in the Roof of Coal Mine A. pdf > <![CDATA[General Survey and Treatment Report on Hidden Disaster-causing Factors of Mine A. pdf > <![CDATA[Annual Geological Survey and Water Control Work Plan and Implementation Opinions for Mine A.]]> pdf > <![CDATA[Report on the Hydrogeological Type Classification of B Mine pdf > <![CDATA[General Survey and Treatment Report on Hidden Disaster-causing Factors of B Mine. pdf > <![CDATA[Report on the Detection Results of the Water-richness of the Roof and Goaf in the Crosshaulage Roadway of the 226 Upper 02 Working Face in Coal Mine B. pdf > <![CDATA[B Mine Geological Environment and Land Reclamation Plan. pdf > <![CDATA[Study on the Mine Pressure Behavior and Overlying Strata Movement Law of the First Mining Face in Coal Seams 4 and 6 of B Mine. pdf > <![CDATA[Report on the Classification of Hydrogeological Types of C Mine pdf > <![CDATA[General Survey and Treatment Report on Hidden Disasters Caused by Coal Mine C pdf >

[0106] Step 2: Find the corresponding report in the entire report set according to {"A Mine Hidden Disaster Factor Census and Governance Report", "B Mine Hidden Disaster Factor Census and Governance Report", "C Mine Hidden Disaster Factor Census and Governance Report"}. Table 2 represents the real report file, not just the name text. Get {A Mine Hidden Disaster Factor Census and Governance Report.pdf, B Mine Hidden Disaster Factor Census and Governance Report.pdf, C Mine Hidden Disaster Factor Census and Governance Report.pdf}.

[0107] Step 3: Get the first report in {A mine hidden disaster factor census and governance report.pdf, B mine hidden disaster factor census and governance report.pdf, C mine hidden disaster factor census and governance report.pdf}, the A mine hidden disaster factor census and governance report. Clean the complex symbols in the A mine hidden disaster factor census and governance report, such as special geological age symbols and unit water inflow units. Segment the entire text in the A mine hidden disaster factor census and governance report according to the sentence structure of the report text, and return the geological sentence set (GSS) of this report. The set content is shown in Table 3.

[0108] Table 3

[0109]

[0110] Step 4: Get the first sentence in GSS: "The three-way gate, broken zone, and water sprinkling area of ​​the excavation tunnel may cause local roof collapse if the roof integrity, stability, anchorage and other factors are not reinforced in time." Identify and extract all words and merge them into GWS, as shown in Table 4.

[0111] Table 4

[0112]

[0113] Step 5: Incorporate {'excavation', 'tunnel', 'fork', 'gate', 'broken', 'with water', 'area', 'roof', 'integrity', 'stability', 'anchorage', 'factor', 'reinforcement', 'support', 'not', 'timely', 'possible', 'cause', 'local', 'roof collapse'} into AGWS. The contents of AGWS are as follows: Figure 8 shown.

[0114] Step 6: Repeat steps 4-5. Figure 8 As shown, the construction of AGWS is realized.

[0115] Step 7: Segment the words "water accumulation in goaf" to obtain the disaster-causing query word set (HKEYS), namely {'goaf', 'water accumulation'}.

[0116] Step 8: Set the parameters of the algorithm. k1, k2, and b are coordination factors, which are set to 2, 1, and 0.75 respectively.

[0117] Step 9: Calculate the similarity between AGWS and {'goaf', 'water accumulation'}. n represents the total number of words in HKEYS, that is, 2, W i Indicates the weight of {'goaf', 'water'}, R (HKEYS i , GSS j ) represents the correlation score between the i-th word in HKEYS and a sentence in GSS. N represents the total number of GSS statements, df i Indicates the total number of statements in the geological report containing {'mined-out area', 'water accumulation'}. in f i Indicates the total number of times {'goaf', 'water accumulation'} appears in a sentence in a geological report. dl indicates GSS j The length of the statement. avg_dl represents the average length of the GSS statement.

[0118] Step 10: Define the Similarity Set (SS).

[0119] Step 11: Add the similarity score to the similarity set.

[0120] Table 5

[0121]

[0122] Step 12: Repeat step 11.

[0123] Step 13: Sort the SS and obtain the corresponding first five geological statements in the GSS.

[0124] Table 6

[0125]

[0126] 2. If Figure 2 As shown, the dynamic generation of report options is realized

[0127] Step 1: Extract the set of all report names (RNS) from the set of all reports (REPORT).

[0128] Table 7

[0129]

[0130] Step 2: Load the RNS set into the hidden disaster factor mining device.

[0131] 3. If Figure 3 As shown, the calculation and implementation of the total word frequency statistics chart

[0132] Step 1: Get the report name set (RNS) to be counted. Create the result word set (WS) and the result word frequency set (WF). Create the mapping relationship f of the set: WS->WF.

[0133] Table 8

[0134] The name set of reports to be counted RNS "Report on the Investigation and Control of Hidden Disaster-causing Factors in Mine A" "Report on the Investigation and Control of Hidden Disaster-causing Factors in Mine B" "Report on the Investigation and Control of Hidden Disaster-causing Factors in Mine C"

[0135] Table 9

[0136] Result words Mapping Results word frequency gas -＞ 621 Fault -＞ 541 Goaf -＞ 443 roof -＞ 380 Water accumulation -＞ 308 Aquifer -＞ 292 … …

[0137] Step 2: Obtain the report set (RS) to be counted from the entire report set (REPORT) according to RNS.

[0138] Table 10

[0139]

[0140] Step 3: Get the i-th report RS in RS i Extract all text content.

[0141] Table 11

[0142]

[0143] Step 4: Add the text content extracted from RS1, RS2, and RS3 to the entire text set (ATEXT).

[0144] Table 12

[0145]

[0146] Step 5: Introduce the stop word set (SWORDS) and the target word set (TWORDS).

[0147] Table 13

[0148]

[0149] Table 14

[0150]

[0151] Step 6: Store each row in ATEXT as a row set (ATEXTLINE).

[0152] Step 7: Extract ATEXTLINE i All words in SWORDS, if the word does not belong to SWORDS j , and the word belongs to TWORDS j , the word is added to WSk The corresponding WF under f k Increase the value by 1.

[0153] Table 15

[0154] Words Is it SWORDS? Is it TWORDS? WS WF … gas no yes gas 1 … Goaf no yes Goaf 1 … therefore yes no Not Join 0 … so yes no Not Join 0 … gas yes no gas 2 … gas yes no gas 3 … Second plate area no no Not Join 0 … Goaf yes no Goaf 2 …

[0155] Step 8: Repeat step 7.

[0156] Step 9: Repeat steps 8-9 until ATEXTLINE is exhausted.

[0157] Step 10: Refer to the top 20 with the largest number in WF. According to f, extract the corresponding WS subset and WF subset.

[0158] Table 16

[0159] <![CDATA[ ws ]]> <![CDATA[ wf ]]> gas 621 Fault 541 Goaf 443 roof 380 Water accumulation 308 Aquifer 292 structure 273 Downhole 210 Base Plate 210 Laneway 196 … …

[0160] 4. If Figure 4 As shown, calculate the subword frequency statistics

[0161] Step 1: Get the set of report names to be counted {A mine hidden disaster factor census and governance report, B mine hidden disaster factor census and governance report, C mine hidden disaster factor census and governance report}. Establish the result word set (WS) and the result word frequency set (WF). Establish the mapping relationship f of the set: WS->WF.

[0162] Table 17

[0163] The name set of reports to be counted RNS "Report on the Investigation and Control of Hidden Disaster-causing Factors in Mine A" "Report on the Investigation and Control of Hidden Disaster-causing Factors in Mine B" "Report on the Investigation and Control of Hidden Disaster-causing Factors in Mine C"

[0164] Table 18

[0165]

[0166] Table 19

[0167]

[0168] Table 20

[0169]

[0170] Step 2: According to {A mine hidden disaster factors census and management report, B mine hidden disaster factors census and management report, C mine hidden disaster factors census and management report}, obtain the report set to be counted {A mine hidden disaster factors census and management report.pdf, B mine hidden disaster factors census and management report.pdf, C mine hidden disaster factors census and management report.pdf} in the entire report set (REPORT).

[0171] Table 21

[0172]

[0173] Step 3: Get the hidden disaster-causing factors census and management report of A mine.pdf in {A mine hidden disaster-causing factors census and management report.pdf, B mine hidden disaster-causing factors census and management report.pdf, C mine hidden disaster-causing factors census and management report.pdf} and extract all text contents.

[0174] Step 4: Introduce the stop word set (SWORDS) and the target word set (TWORDS).

[0175] Table 22

[0176]

[0177] Table 23

[0178]

[0179] Step 5: Store each row of {A Mine Hidden Disaster Factors Survey and Control Report.pdf} as a row set (TEXTLINE).

[0180] Step 6: Extract TEXTLINE i All words in SWORDS, if the word does not belong to SWORDS j , belongs to TWORDS j , the word is added to WS k , the corresponding WF under f k Increase the value by 1.

[0181] Table 24

[0182] Words Is it SWORDS? Is it TWORDS? WS WF … roof no yes gas 1 … Fault no yes Goaf 1 … besides yes no Not Join 0 … at the same time yes no Not Join 0 … roof yes no gas 2 … roof yes no gas 3 … no no Not Join 0 … Fault yes no Goaf 2 …

[0183] Step 7: Repeat step 6 until all words have been cycled through.

[0184] Step 8: Repeat steps 3-7 until the statistics of {B Mine Hidden Disaster Factors Survey and Control Report.pdf, C Mine Hidden Disaster Factors Survey and Control Report.pdf} are completed.

[0185] Step 9: Refer to the top 20 with the largest number in WF. According to f, extract the corresponding WS subset and WF subset.

[0186] Table 25

[0187]

[0188] Table 26

[0189]

[0190] Table 27

[0191]

[0192] 5. If Figure 5 As shown, the front-end browsing report

[0193] Step 1: Get the name of the report to be browsed {"A Mine Hidden Disaster Factor Survey and Control Report"}.

[0194] Step 2: According to {“A Mine Hidden Disaster Factors Survey and Control Report”}, find the same vector in the entire report name set (RNS).

[0195] Table 28

[0196]

[0197] Step 3: According to f: A mine hydrogeological type classification report.pdf->"C: / RNS / A mine hidden disaster factors survey and control report.pdf".

[0198] Step 4: The device accesses the relative path where the local report is stored according to "C: / RNS / A Mine Hidden Disaster Factors Survey and Control Report.pdf" and browses and displays it through the device.

[0199] 6. If Figure 6 As shown, the filter display of the front-end report

[0200] Step 1: Create a reporting mine set (RM) and a reporting category set (RV).

[0201] Table 29

[0202] Report mining area collection RM A mining area B Mining Area C Mining Area …

[0203] Table 30

[0204] Report Category Set RV Hydrogeological classification Geophysical exploration Rules and regulations Ledger …

[0205] Step 2: Create a full report set (REPORT).

[0206] Table 31

[0207] General Report Collection <![CDATA[Report on the Hydrogeological Type Classification of Mine A. pdf > <![CDATA[Mine Geological Environment Protection and Land Reclamation Plan. pdf > <![CDATA[3D Geological Exploration Report of the East Wing of A Coal Mine Wellfield. pdf > <![CDATA[Transient Electromagnetic Exploration Report of 2102 Working Face in A Coal Mine. pdf > <![CDATA[Research Report on the Detection of the "Three Zones" Height in the 3101 Working Face of A Coal Mine. pdf > <![CDATA[Summary Report on Detection of Wide Falling Fissure Zone in Roof of Coal Mine A. pdf > <![CDATA[General Survey and Treatment Report on Hidden Disaster-causing Factors of Mine A. pdf > <![CDATA[Annual Geological Survey and Water Control Work Plan and Implementation Opinions for Mine A.]]> pdf > <![CDATA[Report on the classification of hydrogeological types of B ore deposit. pdf > <![CDATA[General Survey and Treatment Report on Hidden Disaster-causing Factors of B Mine. pdf > <![CDATA[Report on the Detection Results of the Water Richness of the Roof and Goaf in the 226 Upper 02 Working Face Cross-haulage Roadway of B Coal Mine. pdf > <![CDATA[B Mine Geological Environment and Land Reclamation Plan. pdf > <![CDATA[Study on the Mine Pressure Behavior and Overlying Strata Movement Law of the First Mining Face in Coal Seams 4 and 6 of B Mine pdf > <![CDATA[Report on the Classification of Hydrogeological Types of C Mine. pdf > <![CDATA[General Survey and Treatment Report on Hidden Disasters Caused by Coal Mine C pdf > …

[0208] Step 3: According to the mapping relationship f: A mining area, geophysical prospecting report->{A coal mine field east wing three-dimensional geological exploration report.pdf, A coal mine 3101 working face "three belts" height detection research report.pdf, A coal mine 3101 working face "three belts" height detection research report.pdf, A coal mine roof wide fracture zone detection summary report.pdf}

[0209] Step 4: Based on {3D geological exploration report on the east wing of A coal mine field.pdf, “three zones” height detection research report on 3101 working face of A coal mine.pdf, “three zones” height detection research report on 3101 working face of A coal mine.pdf, “summary report on detection of wide-falling fracture zone in roof of A coal mine.pdf}, establish the name subset {“3D geological exploration report on the east wing of A coal mine field”, “Research report on “three zones” height detection research report on 3101 working face of A coal mine”, “Research report on “three zones” height detection research report on 3101 working face of A coal mine”, “Summary report on detection of wide-falling fracture zone in roof of A coal mine”}.

[0210] Step 5: The device uses {“Three-dimensional geological exploration report on the east wing of A coal mine field”, “Three-zone height detection research report on 3101 working face of A coal mine”, “Three-zone height detection research report on 3101 working face of A coal mine”, “Summary report on detection of wide-fall fracture zone in roof of A coal mine”} to display all file names after filtering categories.

[0211] 2. The device is as shown in the instruction manual. Figure 7 shown.

[0212] 1-Report selection area. This diagram shows: report data set preview and selection, pull down the multiple-select list to check the report. Check in the manual: coal mine hidden disaster factor survey report, working face transient electromagnetic exploration report, roof span fracture zone detection summary report, mine geological environment and land reclamation plan.

[0213] 2- Fuzzy search area. Enter key information to quickly retrieve relevant content in the data set. For example, fuzzy match "water-conducting fracture zone threat" with all statements in the four checked reports to obtain the content in each report that is most relevant to "water-conducting fracture zone threat".

[0214] 3-Total word frequency statistics area. Calculate and display the word frequency statistics bar chart in the hidden disaster factor report, with the horizontal axis being the frequency of the word and the vertical axis being the word.

[0215] 4-Total word cloud area. Calculate and display the total word frequency. The higher the word frequency, the larger the font size of the word.

[0216] 5, 6, 7, 8 - Sub-word frequency statistics area. Calculate and display the sub-word frequency statistics of multiple reports. This figure shows: word frequency statistics of coal mine hidden disaster factors survey report, word frequency statistics of working face transient electromagnetic exploration report, word frequency statistics of roof span fracture zone detection summary report, word frequency statistics of mine geological environment and land reclamation plan. Used for comparison between the whole and part, part and part.

[0217] 5, 6, 7, 8 - Sub-word cloud area. Calculate and display sub-word cloud of multiple reports. This diagram shows: word cloud of coal mine hidden disaster factors survey report, word cloud of working face transient electromagnetic exploration report, word cloud of roof span fracture zone detection summary report, word cloud of mine geological environment and land reclamation plan. The higher the word frequency, the larger the font size of the word.

[0218] 9- Screening display area. Report screening module, combined with the reporting system for quick screening.

[0219] 10- Report browsing area. Display detailed content for quick browsing.

[0220] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed in the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims. The information disclosed in the background technology section of this article is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as an admission or in any form that the information constitutes prior art known to those skilled in the art.

Claims

1. A method for automatically identifying hidden disaster-causing factors in mines based on text mining, characterized in that: include: Keyword-based matching of hidden disaster-causing factors information; Implement dynamic generation of report options; Calculation and implementation of total word frequency statistics; Calculate subword frequency statistics; Front-end browsing report; Filtered display of front-end reports.

2. The method for automatically identifying hidden disaster-causing factors in mines based on text mining according to claim 1 is characterized in that: The keyword-based hidden disaster factor information matching includes: Definition: Establish the Geological Word Set (GWS) and the General Geological Word Set (AGWS); Step 1: Obtain the report name set (RNS) to be queried and the hidden disaster-causing key factors (HKEY) to be queried; Step 2: According to the report name set (RNS), obtain the report set (RS) to be queried from the entire report set (REPORT); Step 3: Get the i-th report RS in the report set (RS) i Clean complex symbols in the report, such as special geological age symbols, unit water inflow units, etc., and replace RS i Split the report text according to its sentence structure and return the geological sentence set (GSS) of the report. Step 4: Get the jth statement GSS in the geological statement set (GSS) j , identify and extract GSS j All words in the Geological Word Set (GWS) are merged into the Geological Word Set (GWS); Step 5: Get the kth set GWS in GWS k , incorporated into the General Set of Geological Words (AGWS); Step 6: Repeat steps 4-5; Step 7: Process the hidden disaster-causing key factors (HKEY) as in steps 4-6 to obtain a disaster-causing query word set (HKEYS); Step 8: Set the parameters of the algorithm. k1, k2, and b are coordination factors, which are set to 2, 1, and 0.75 respectively. Step 9: Calculate the similarity between the total set of geological words (AGWS) and the hidden disaster-causing key factors (HKEY): n represents the total number of words in HKEYS, W i Indicates HKEYS i The weight owned, R(HKEYS i , GSS j ) represents the relevance score between the i-th word in HKEYS and a sentence in the geological sentence set (GSS), where N represents the total number of statements in the geological statement set (GSS), df i Indicates that HKEYS is included i The total number of sentences in the geological report, of which in f i Indicates HKEYS i The total number of times a sentence appears in a geological report. dl represents GSS j The length of the statement, avg_dl represents the average length of the Geological Statement Set (GSS) statements; Step 10: Define the similarity set (SS); Step 11: Add the similarity score to the similarity set; Step 12: Repeat step 11; Step 13: Sort the similarity set (SS) to obtain the first five corresponding geological statements in the geological statement set (GSS).

3. The method for automatically identifying hidden disaster-causing factors in mines based on text mining according to claim 2 is characterized in that: The dynamic generation of report options includes: Step 1: Extract the set of all report names (RNS) from the set of all reports (REPORT); Step 2: Load all report name sets (RNS) into the hidden disaster factor mining device.

4. The method for automatically identifying hidden disaster-causing factors in mines based on text mining according to claim 3 is characterized in that: The calculation and implementation of the total word frequency statistics include: Step 1: Get the report name set (RNS) to be counted, establish the result word set (WS), establish the result word frequency set (WF), and establish the mapping relationship f of the set: WS->WF; Step 2: Obtain the report set (RS) to be counted from the entire report set (REPORT) according to the report name set (RNS); Step 3: Get the i-th report RS in the report set (RS) i Extract all text content; Step 4: Add the text content extracted in step 3 to the total text collection (ATEXT); Step 5: loop steps 3-4 until all text contents in the report set (RS) are added to the total text set (ATEXT); Step 6: Introduce the stop word set (SWORDS) and the target word set (TWORDS); Step 7: Store each line in the entire text set (ATEXT) as a line set (ATEXTLINE); Step 8: Extract ATEXTLINE i All words in SWORDS, if the word does not belong to SWORDS j , and the word belongs to TWORDS j , the word is added to WS k The corresponding WF under the mapping relationship f k Add 1 to the value; Step 9: Repeat step 8; Step 10: Loop steps 8-9 until the row set (ATEXTLINE) is exhausted; Step 11: Referring to the top 20 largest words in the result word frequency set (WF), according to the mapping relationship f, extract the corresponding result word set (WS) subset and result word frequency set (WF) subset.

5. The method for automatically identifying hidden disaster-causing factors in mines based on text mining according to claim 4 is characterized in that: The calculation of subword frequency statistics includes: Step 1: Get the report name set (RNS) to be counted, establish the result word set (WS), establish the result word frequency set (WF), and establish the mapping relationship f of the set: WS->WF; Step 2: Obtain the report set (RS) to be counted from the entire report set (REPORT) according to the report name set (RNS); Step 3: Get the i-th report RS in the report set (RS) i Extract all text content; Step 4: Introduce the stop word set (SWORDS) and the target word set (TWORDS); Step 5: RS i Each row is stored as a row set (TEXTLINE); Step 6: Extract TEXTLINE i All words in SWORDS, if the word does not belong to SWORDS j , belongs to TWORDS j , the word is added to ws k , the corresponding WF under f k Add 1 to the value; Step 7: Repeat step 6; Step 8: Repeat steps 3-7 until the report set (RS) is exhausted; Step 9: Referring to the top 20 with the largest number in WF, according to the mapping relationship f, extract the corresponding result word set (WS) subset and result word frequency set (WF) subset.

6. The method for automatically identifying hidden disaster-causing factors in mines based on text mining according to claim 5 is characterized in that: The front-end browsing report includes: Step 1: Get the report name (RN) to be browsed and define the report path set (RPS); Step 2: According to RN, find the same vector in the entire report name set (RNS); Step 3: According to f:RNS i ->RPS i , find the path to which the corresponding report name is mapped; Step 4: Install according to RPS i Display the report content.

7. The method for automatically identifying hidden disaster-causing factors in mines based on text mining according to claim 6 is characterized in that: The screening display of the front-end report includes: Step 1: Create a reporting mine set (RM) and a reporting category set (RV); Step 2: Create a set of all reports (REPORT); Step 3: According to the mapping relationship f:RM i ,RV j ->REPORT k , get the report subset (RSUB); Step 4: Create a report name subset (RNSUB) based on RSUB; Step 5: The device displays all file names after the filter category using Report Name Subset (RNSUB).

Images