A construction project list data processing method, engineering quantity pricing method and device
By extracting and processing the feature items and keywords in the construction project list, and establishing professional vocabulary dictionaries and regular expressions, the problem of low matching rate in the existing technology is solved, and more accurate project quantity metering and intelligent price grouping recommendations are achieved.
Patent Information
- Application Number
- CN202110973597.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-08-24
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2041-08-24
AI Technical Summary
The existing intelligent price recommendation quota group method is difficult to accurately match the list description in construction projects. It is mainly because the common word segmentation tool cannot extract special vocabulary from the construction industry and divide the key words related to the quota, which affects the matching rate and pricing results. accuracy.
By obtaining the list description of historical building projects, extracting feature items and keywords, establishing professional vocabulary dictionaries and regular expressions, building metadata collections, performing structured processing to generate a list description of the project that meets the field of construction, and using this description for quorum groups match.
It improves the performance of intelligent construction engineering products and the accuracy of project quantity valuation results, ensures that the list description only contains professional vocabulary and expressions related to the quota, and enhances the accuracy of matching results.
Smart Images

Figure CN113806392B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of engineering construction, and in particular to a method for processing construction engineering list data, a method and a device for calculating the engineering quantity. Background Art
[0002] In the intelligent technologies related to construction projects, such as the intelligent pricing products used to realize the engineering quantity pricing, they all need to be based on the data in the construction project list. Taking the pricing of construction projects as an example, the use of the intelligent pricing function to recommend quota groups is to serve the construction project price. The engineering quantity pricing is to determine the quota group based on the specific engineering content described in the engineering list; the selected quota group has the engineering quantity price, and the final sum of the engineering quantity prices of the quota group determines the pricing of the entire project. The quota pricing is based on the given engineering list, according to the list data, correctly apply the project quotas included in the list, and then multiply the engineering quantity by the quota unit price to calculate the total price of the entire project.
[0003] The existing intelligent pricing method for recommending quota groups is to find the most matching list description from the historical list description, and then recommend the quota group with the matching list description as the quota combination plan. Whether it matches or not is determined by the word segmentation similarity of the list description, and the accuracy of the similarity depends on the word segmentation result. However, the existing word segmentation tools are mostly general dictionaries that cannot extract specialized vocabulary for the construction industry. At the same time, the word segmentation is also general, and a large number of words are segmented, and most of them have nothing to do with the quota, which affects the final matching rate.
[0004] Therefore, how to accurately and reasonably describe the construction project list is of great significance to the application of intelligent technology in the field of construction engineering. Summary of the invention
[0005] In view of this, an embodiment of the present invention provides a method and device for describing construction project list data to solve the problem that the existing technology of using word segmentation tools to describe construction project lists is difficult to meet the application needs of intelligent technology in the construction industry.
[0006] According to a first aspect, an embodiment of the present invention provides a method for describing construction project list data, comprising:
[0007] Obtain a collection of inventory descriptions of historic buildings and inventory processing requirements;
[0008] Extracting feature items from the historical building project inventory description set based on the inventory processing requirements to obtain a feature item set;
[0009] Determine the feature value category based on the keyword corresponding to each feature item in the feature item set, and establish a corresponding feature value professional vocabulary dictionary and feature value regular expression, wherein the feature value professional vocabulary dictionary and feature value regular expression are used to extract professional vocabulary corresponding to the feature value category;
[0010] Based on the feature item set and the feature value category of each feature item and its corresponding feature value professional vocabulary dictionary and feature value regular expression, a metadata set is formed;
[0011] The project content of the pending construction project list is processed based on the metadata set to obtain a structured project list description.
[0012] Optionally, the feature item extraction is performed on the historical building engineering inventory description set based on the inventory processing requirement to obtain a feature item set, including:
[0013] Determine a feature item identifier corresponding to the feature item name based on the list processing requirement;
[0014] Extracting feature item names from the historical building engineering inventory description set based on the feature item identifier;
[0015] The feature item set is constructed based on the extracted feature item names.
[0016] Optionally, determining the feature value category based on the keyword corresponding to each feature item in the feature item set, and establishing a corresponding feature value professional vocabulary dictionary and feature value regular expression, includes:
[0017] Extract keywords from the content corresponding to the current feature item, the keywords include: text keywords and numeric keywords;
[0018] Define the category names of the extracted keywords and determine the feature value categories;
[0019] A corresponding feature value professional vocabulary dictionary is established based on text keywords, and a corresponding feature value regular expression is established based on numeric keywords.
[0020] Optionally, the processing of the project content of the pending construction project list based on the metadata set to obtain a structured project list description includes:
[0021] Based on the feature value professional vocabulary dictionary and feature value regular expression in the metadata set, feature extraction is performed on the engineering content of the pending construction engineering list to obtain corresponding professional vocabulary;
[0022] The extracted professional vocabulary is classified into characteristic value categories and characteristic items according to its corresponding characteristic value professional vocabulary dictionary or characteristic value regular expression to generate a structured engineering list description.
[0023] Optionally, the method further comprises:
[0024] Obtaining a list of several historic building works;
[0025] Based on the area code, list guide version and list code corresponding to the list of building projects to be processed, the description of the list of historical building projects is screened to determine the set of descriptions of the list of historical building projects.
[0026] According to a second aspect, an embodiment of the present invention provides a method for calculating engineering quantities, including:
[0027] Obtain a list of construction works to be priced and a collection of historical reference construction work lists;
[0028] The list processing requirement is set to quota selection, and the method of the first aspect or any corresponding implementation method thereof is adopted to process the to-be-priced construction project list and the historical reference construction project list respectively, to obtain a corresponding to-be-priced construction project list description and a historical reference construction project list description set;
[0029] Calculating the matching degree between the to-be-priced engineering list description and each historical reference engineering list description in the historical reference engineering list description set respectively;
[0030] Obtain the historical quota group corresponding to the historical reference engineering list description with the highest matching degree;
[0031] The construction work list to be priced is priced based on the historical quota group.
[0032] According to a third aspect, an embodiment of the present invention provides a construction project list data processing device, comprising:
[0033] The first acquisition module is used to obtain a collection of historical building project inventory descriptions and inventory processing requirements;
[0034] A first processing module is used to extract feature items from the historical building project inventory description set based on the inventory processing requirements to obtain a feature item set;
[0035] A second processing module is used to determine the feature value category based on the keyword corresponding to each feature item in the feature item set, and to establish a corresponding feature value professional vocabulary dictionary and feature value regular expression, wherein the feature value professional vocabulary dictionary and feature value regular expression are used to extract professional vocabulary corresponding to the feature value category;
[0036] A third processing module, configured to construct a metadata set based on the feature item set and the feature value category of each feature item and its corresponding feature value professional vocabulary dictionary and feature value regular expression;
[0037] The fourth processing module is used to process the project content of the pending construction project list based on the metadata set to obtain a structured project list description.
[0038] According to a fourth aspect, an embodiment of the present invention provides a device for calculating engineering quantity, including:
[0039] The second acquisition module is used to obtain a list of construction projects to be priced and a collection of historical reference construction project lists;
[0040] A fifth processing module is used to set the list processing requirement to quota selection, and use the construction project list data processing device provided by another embodiment of the present invention to process the to-be-priced construction project list and the historical reference construction project list respectively to obtain a corresponding to-be-priced construction project list description and a historical reference construction project list description set;
[0041] A sixth processing module, used to respectively calculate the matching degree between the to-be-priced engineering list description and each historical reference engineering list description in the historical reference engineering list description set;
[0042] The seventh processing module is used to obtain the historical quota group corresponding to the historical reference engineering list description with the highest matching degree;
[0043] The eighth processing module is used to perform engineering quantity pricing on the list of construction works to be priced based on the historical quota group.
[0044] According to the fifth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores computer instructions, and when the computer instructions are executed by a processor, the method described in the first aspect of the present invention and any optional method thereof is implemented.
[0045] According to the sixth aspect, an embodiment of the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method described in the first aspect of the present invention and any optional method thereof by executing the computer instructions.
[0046] The technical solution of the present invention has the following advantages:
[0047] 1. The embodiment of the present invention provides a method and device for processing building project list data, which obtains a historical building project list description set and a list processing requirement; extracts feature items from the historical building project list description set based on the list processing requirement to obtain a feature item set; determines a feature value category based on the keyword corresponding to each feature item in the feature item set, and establishes a corresponding feature value professional vocabulary dictionary and feature value regular expression; forms a metadata set based on the feature item set and the feature value category of each feature item and its corresponding feature value professional vocabulary dictionary and feature value regular expression; processes the engineering content of the building project list to be processed based on the metadata set to obtain a structured engineering list description. Thus, by determining feature items from the historical building project list description, and determining feature categories according to the content of the feature items to establish corresponding feature value professional vocabulary dictionary and feature value regular expression, professional vocabulary and expressions that conform to the construction field can be segmented, and by structurally classifying the segmented professional vocabulary and expressions into feature items according to the technical meaning of the engineering content of the list, and only extracting key words associated with the list processing requirement, the final engineering list description is more in line with the intelligent technology requirements, which is conducive to improving the performance of intelligent products for building projects.
[0048] 2. The embodiment of the present invention provides a method and device for engineering quantity pricing, which obtains a list of construction projects to be priced and a set of historical reference list of construction projects; sets the list processing requirement to be selected as quota, and uses the construction project list data processing method provided by another embodiment of the present invention to process the list of construction projects to be priced and the historical reference list of construction projects respectively, and obtain the corresponding list description of the construction projects to be priced and the set of historical reference list descriptions; respectively calculates the matching degree of the list description of the construction projects to be priced and each historical reference list description in the set of historical reference list descriptions; obtains the historical quota group corresponding to the historical reference list description with the highest matching degree; and performs engineering quantity pricing on the list of construction projects to be priced based on the historical quota group. Thus, by using the description method provided by another embodiment of the present invention to perform structured description on the list of construction projects to be priced and the historical reference list of construction projects, the obtained list description only includes the content related to the quota, and the list description is the vocabulary and expression of the construction profession, and the list description is used to match the quota group, which can effectively improve the accuracy of the matching result, and thus improve the accuracy of the engineering quantity pricing result. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0050] Figure 1 A schematic diagram of a process of recommending a quota group by an existing intelligent price group in an embodiment of the present invention;
[0051] Figure 2 Flow chart of a method for processing construction project list data in an embodiment of the present invention;
[0052] Figure 3 Flow chart of the method for calculating the engineering quantity in an embodiment of the present invention;
[0053] Figure 4 It is a structural schematic diagram of a construction project list data processing device in an embodiment of the present invention;
[0054] Figure 5 It is a structural schematic diagram of the engineering quantity pricing device in an embodiment of the present invention;
[0055] Figure 6 Schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION
[0056] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, 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.
[0057] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0058] In the embodiment of the present invention, the intelligent pricing function in the construction engineering software is taken as an example to describe in detail the construction engineering list data processing method, engineering quantity pricing method and device provided by the embodiment of the present invention.
[0059] First, the basic principles of the existing method of intelligent pricing and quota group recommendation are introduced as follows:
[0060] The basic principle of the existing intelligent pricing and quota group recommendation technology method includes the steps of word segmentation of the list description, calculating the similarity of the list description, finding the most similar historical list description, and the quota group it carries is the required quota group. Among them, the quality of word segmentation has a great influence on the matching rate and accuracy of the recommendation. The specific process includes preprocessing of the list description, text segmentation, calculating the similarity of the word segmentation string and the text segmentation string of the historical list description, and recommending the quota group with the highest similarity.
[0061] like Figure 1 As shown, the specific steps are a) preprocessing the description of the list; b) text segmentation process; c) calculating the similarity of sentences; d) recommending the historical list quota group that is most similar to the list description.
[0062] a) Preprocessing of the description of the list
[0063] The list description is a description of the characteristics of the project. It is used to express the essential content of the list items. It will directly determine the value of the project itself. Therefore, the accuracy of the description of the characteristics of the bill of quantities items is directly related to the accuracy of the comprehensive unit price. The list description used in the intelligent price combination is processed as follows:
[0064] Example of a manifest description - the manifest includes a description of the manifest to match, and a collection of manifest descriptions from historical project files.
[0065] Example of list description: 1. Prestressed slab; 2. Ordinary commercial concrete crushed stone particle size 20 stone C20; 3. Flat plate, trough plate within 5m hollow plate; 4. Cement mortar M5. Preliminary processing of the list description: for example: unification of symbols, regularization of characters '.'=>'.','m 2 '=>'square meters', 'm 3 '=>'cubic meter'.
[0066] b) Text segmentation process
[0067] The word segmentation process is to combine the continuous Chinese character sequence described in the list into a new word sequence according to certain specifications, that is, to divide it into individual word sequences. The Chinese word segmentation algorithm is much more complicated than the Latin of the European and American writing system. English words are usually separated by spaces, and word segmentation can be grouped by spaces or other separators, but Chinese does not have an obvious separator or no separator at all, so Chinese word segmentation is somewhat difficult.
[0068] There are many Chinese word segmentation algorithms. The most common one is the dictionary-based word segmentation algorithm, which matches the text sentence with a large dictionary. The successfully matched string is the segmented phrase. Of course, the actual word segmentation algorithm will be more complicated. For example, the size of the dictionary, the source of the dictionary vocabulary, and the matching algorithm will affect the accuracy and efficiency of word segmentation.
[0069] The most commonly used Chinese word segmentation tool is the open source Jieba, which uses a dictionary-based word segmentation algorithm. Other word segmentation tools include Stanford Word Segmenter and HanLP, which is also a good tool.
[0070] Participle example:
[0071] List description 1: "Straight wall; 1. Concrete strength grade: ordinary commercial concrete crushed stone particle size 20 stone C50; 2. Buildings that are too high will increase labor and reduce mechanical efficiency. The height is within 90m."
[0072] The word segmentation results are:
[0073] ['straight','shape','wall',';','concrete','strength','grade','common commodity','concrete','gravel','particle size','20','stone','C50','building','super high','increase','manual','machinery','reduce','efficiency','height','90m','within'], using the "Jieba word segmentation tool", deleted punctuation marks and some useless characters ('1.','2.','.',';',':','、').
[0074] Listing description 2: "Straight wall; 1. Concrete strength grade: ordinary commercial concrete crushed stone particle size 20 stone C30".
[0075] The word segmentation results are:
[0076] ['straight', 'shape', 'wall', 'concrete', 'strength', 'grade', 'common commodity', 'concrete', 'gravel', 'particle size', '20', 'stone', 'C30'], using the "Jieba word segmentation tool", deleted punctuation marks and some useless characters ('1.', '2.', '.', ';', ':', '、').
[0077] List description 3: "1. Straight wall; 2. Concrete strength grade: C50 commercial concrete; 3. Contains hidden columns and connecting beams."
[0078] The word segmentation results are:
[0079] ['straight', 'shape', 'wall', 'merge', 'production', 'sub-item', 'commodity', 'underwater', 'concrete', 'gravel', 'particle size', '20', 'stone', 'C50'], using the "Jieba Word Segmentation Tool", deleted punctuation marks and some useless characters ('1.', '2.', '.', ';', ':', '、').
[0080] Listing description 4: "1. Vertical wall combined production sub-item product underwater concrete crushed stone particle size 20 stone C30".
[0081] The word segmentation results are:
[0082] ['straight', 'shaped', 'wall', 'merge', 'production', 'sub-item', 'commodity', 'underwater', 'concrete', 'gravel', 'particle size', '20', 'stone', 'C30'], using the "Jieba Word Segmentation Tool", deleted punctuation marks and some useless characters ('1.', '2.', '.', ';', ':', '、').
[0083] c) Calculate the similarity of sentences
[0084] By calculating the similarity of the list descriptions, we can find examples of similar historical project list descriptions. The examples will contain quota price information, so that the list price can be calculated.
[0085] There are many ways to calculate sentence similarity. The most common method is the "edit distance calculation" (Levenshteint Distance), which is to calculate the minimum number of edits required to transform two strings from one to another, that is, the minimum distance. The number of edits can be defined as the replacement, addition, or deletion of a character. The greater the distance between the two strings, the more different they are, and the lower the similarity. Taking the "edit distance" algorithm as an example, first calculate the sentence "edit distance" (Levenshteint Distance) between sentences. The results of the algorithm are listed in Table 1.
[0086] Table 1
[0087]
[0088] The "edit distance" of a sentence refers to the number of edits required to convert a string of one sentence into a string of another sentence (that is, the strings of two sentences). The fewer the number, the more similar the strings of the two sentences are. Taking the list description in Table 1 as an example, the most similar list description can be obtained, as shown in Table 2. It can be seen that the similarity calculation depends on the length of the two strings and how many characters are in common.
[0089] Table 2
[0090]
[0091]
[0092] d) Recommend the historical list quota group that is most similar to the list description
[0093] Recommending a list quota group is to find the list description that is most similar / matching to the required list description from the historical list library. The quota group with the matching list description is the recommended quota group. Historical list descriptions generally have quota groups, so as long as the list descriptions match, a suitable quota group can be found. For example, the quota groups with the above list descriptions are shown in Table 3. Then, based on the calculated similarity of the list descriptions, the most similar list description in history is selected, and then the quota group of the most similar list description is promoted as the quota group of the required list description.
[0094] Table 3
[0095]
[0096] In summary, the existing intelligent pricing method for recommending quota groups is to find the most matching list description from the historical list description, and then recommend the quota group with the matched list description as the quota combination plan. Whether it matches or not is determined by the word segmentation similarity of the list description, and the accuracy of the similarity depends on the word segmentation result. The existing word segmentation tools use very general dictionaries and cannot extract specialized vocabulary for the construction industry. At the same time, the word segmentation is also general. Segmenting too many words (most of which have nothing to do with the quota) affects the matching rate.
[0097] It can be seen that the key to improving the accuracy of the quota group recommended by the intelligent group is to find the most matching list, so how to accurately select the most similar list becomes an urgent problem to be solved. The existing method of using general word segmentation tools affects the matching rate and accuracy of the recommended quota group in the following aspects:
[0098] 1. List projects in the construction field contain many words with unique professional characteristics. These professional words are difficult to extract through general word segmentation tools, so it is necessary to develop new tools that can extract special words in the construction field;
[0099] 2. The list describes the characteristics of the project, and the quota of the list is determined by the content of the characteristics of the list. The existing word segmentation tools cannot classify the characteristic values in the list description in a structured manner, and cannot put words with the same characteristic content together. Using structured and classified professional vocabulary to match the list description with the same professional vocabulary, the project content will be more consistent with the project quantity and price represented by the quota, and it will be more accurate.
[0100] Therefore, how to classify the professional vocabulary of the list description according to the project characteristics is the key to correctly matching the quota to improve the accuracy of the intelligent price grouping recommendation quota grouping method. Therefore, a construction project list data processing method is proposed to achieve a structured description of the construction project list to improve the accuracy of the positioning group. Specifically, solutions are given to the following problems of the prior art:
[0101] 1. Using general word segmentation tools cannot extract professional vocabulary in the field of the construction industry
[0102] The existing general word segmentation tools, such as Jieba, IK, Ansj, Hanlp, Stanford, etc., can only segment general words. For special professional words, such as words in the field of architecture, the word segmentation effect is not very good. General word segmentation systems are based on limited training data, or the collected dictionaries are also collected from general articles, such as the words of People's Daily, online novels, etc. It is difficult to capture the language features and unique words applicable to professional fields through limited training data using such general word segmentation tools.
[0103] However, the list description involves specialized vocabulary in the construction industry, especially those that determine the quota group. These specialized vocabulary should be separated into one, rather than divided into common words. For example, the vocabulary in the list description mentioned above:
[0104] List description: "Straight wall; 1. Concrete strength grade: ordinary commercial concrete crushed stone particle size 20 stone C50; 2. Buildings with super-high height increase labor and reduce mechanical efficiency within 90m"
[0105] The word segmentation results are:
[0106] ['straight','shape','wall',';','concrete','strength','grade','ordinary commodity','concrete','crushed stone','particle size','20','stone','C50','building','super height','increase','manual','machinery','reduce','efficiency','height','90m','within'].
[0107] The construction industry's special vocabulary is:
[0108] 'Straight wall', 'Concrete strength grade', 'Ordinary commercial concrete', 'Crushed stone size', '20 stone', 'C50', etc.
[0109] 2. Existing word segmentation tools cannot extract key words related to quotas
[0110] Existing word segmentation tools cannot extract only the key words related to the quota. The words extracted include meaningful words (that is, the key words that can determine the quota group) and meaningless words. Meaningless words will affect the calculation of the similarity of the list description. Dissimilar list descriptions (large edit distance) do not mean that the keywords related to the list description and the quota group are different. Taking the list description provided above as an example, we can see that the results of comparing the lists using the words extracted by the existing word segmentation tools are very inaccurate.
[0111] Example:
[0112] As shown in Table 4, assuming that the quota groups of list description 1 and list description 3 are the same, the similarity of the two list descriptions should be relatively high. The quota group is determined by the comparison of key words, not the comparison of all the separated words. If the key words of the list description are used to match the list, the result will be more accurate.
[0113] Table 4
[0114]
[0115] 3. The matching of the listing description requires characteristic keywords
[0116] The engineering list describes the characteristics of the engineering project. The project characteristics are used to express the essential content of the project name. The content and value of these characteristics determine which appropriate quota group to select. However, the existing word segmentation tools cannot classify the engineering characteristics of the list.
[0117] The list in the construction field contains engineering characteristics, which may contain different categories, and different categories have different values. The keywords of these characteristic item categories will determine the selection of quotas.
[0118] Finding other similar lists based on the list feature items and feature values to determine the quota composition is much more accurate than calculating the similarity of simple word segmentation. For example, as shown in Table 5, the above list description 1 and list description 3 can accurately find similar lists with common quota composition by using the keywords of the structured list description.
[0119] Table 5
[0120]
[0121]
[0122] Based on the above analysis, an embodiment of the present invention provides a method for processing construction project list data, such as Figure 2 As shown, the construction project list data processing method specifically includes the following steps:
[0123] Step S101: Obtain a historical building project inventory description set and inventory processing requirements.
[0124] Among them, in the embodiment of the present invention, the list processing requirements are used as an example for quota selection. The list description requirement is to provide a list description to facilitate the determination of the quota group. In actual applications, the list processing requirements can also be set according to actual needs, and the present invention is not limited to this.
[0125] Step S102: extracting feature items from the historical building project inventory description set based on inventory processing requirements to obtain a feature item set.
[0126] Specifically, a machine learning method is used to extract feature items from a large set of historical building project list descriptions, where the feature items are used to define the project content related to the set of quota groups. For example, the feature item name can be extracted from all descriptions of this list in the historical list data according to special symbols in the feature item name, such as ":".
[0127] Step S103: determining the feature value category based on the keyword corresponding to each feature item in the feature item set, and establishing a corresponding feature value professional vocabulary dictionary and feature value regular expression.
[0128] Among them, the feature value professional vocabulary dictionary and feature value regular expression are used to extract professional vocabulary corresponding to the feature value category. Each feature item includes multiple feature values, and each feature value can be collected using the corresponding professional dictionary (vocabulary) and defined regular expression (pattern). The feature value category is used to structure the extracted professional vocabulary and digital expressions corresponding to the feature value into the feature items of the corresponding list according to the semantic keywords.
[0129] Step S104: construct a metadata set based on the feature item set and the feature value category of each feature item and its corresponding feature value professional vocabulary dictionary and feature value regular expression.
[0130] Specifically, the metadata set mentioned above may also include: rules for standardizing characteristic values, etc., which are used to unify professional vocabulary and numbers into standard expressions according to the characteristics of the inventory, so as to ensure the consistency of descriptions of different project inventories.
[0131] Step S105: Processing the project content of the pending construction project list based on the metadata set to obtain a structured project list description.
[0132] Specifically, by utilizing the above metadata set to extract professional vocabulary and professional numbers from the engineering contents of the construction engineering list to be processed, and classifying the professional vocabulary and professional numbers into feature categories corresponding to corresponding feature items, a structured description of the engineering list is achieved.
[0133] By executing the above steps, the construction project list data processing method provided by the embodiment of the present invention determines the feature items from the historical construction project list descriptions, and determines the feature categories according to the contents of the feature items to establish corresponding feature value professional vocabulary dictionaries and feature value regular expressions, thereby segmenting the professional vocabulary and expressions that conform to the construction field, and structurally classifying the segmented professional vocabulary and expressions into feature items according to the technical meaning of the project content of the list, and only extracting the key words associated with the list processing requirements, so that the final project list description is more in line with the intelligent technology requirements, which is beneficial to improving the performance of intelligent products for construction projects.
[0134] Specifically, in one embodiment, before executing the above step S101, the above construction project list data processing method further includes the following steps:
[0135] Step S106: Obtain a list of several historical building projects.
[0136] Step S107: Based on the region code, list guide version and list code corresponding to the list of building projects to be processed, the historical building project list descriptions are screened to determine a set of historical building project list descriptions.
[0137] Specifically, since there are tens of thousands of checklists in construction projects, it is very unrealistic to manually define the characteristic items and corresponding characteristic values of the checklists. Professional vocabulary and expressions vary a lot in actual projects, and different professionals will use their own familiar and habitual vocabulary and expressions. The description of the project will also have regional differences, and many checklist descriptions are not consistent with national standards. Therefore, it is impossible to define professional vocabulary and expressions based on national and regional checklist standard guidelines in a complete and accurate manner.
[0138] The professional vocabulary and expressions in the list description cannot use a unified national standard for the following reasons: different professionals will use their own familiar and habitual vocabulary and expressions; different regions have different vocabulary and expressions; national standards are not mandatory and self-defined vocabulary and expressions are allowed.
[0139] In order to completely collect the professional vocabulary and expressions in the industry, including different regions and different engineering files, the present invention uses a machine learning-based method to collect the characteristic items of the engineering list, and screens the description of the historical building engineering list according to the regional code, the version of the list guide and the list code, so as to obtain a set of engineering lists consistent with the building engineering list to be processed, and the metadata set is also stored according to the regional code, the version of the list guide and the list code, as shown in Table 6, so that the metadata set can be directly called by subsequent engineering lists of the same type. And it can take into account the uniformity and flexibility of the list description method, so that the description of the engineering list is more in line with the actual situation of the building engineering list.
[0140] Table 6
[0141] area Area Code Listing Guide Version Listing code Inventory data collection ID Guangdong 1710 2013 010504001 00001 Beijing 1714 2013 010540001 00201 Shaanxi 1187 2009 010302003 30245
[0142] Exemplarily, the metadata set in the embodiment of the present invention is based on the regional code, the list guide version, and the nine-digit list code. Different regions, list guide versions and nine-digit list codes are a data set. The main data set includes the feature items of the list, the categories of different values contained in the feature items, the dictionary required to extract the vocabulary corresponding to the feature value and the regular expression of the digital type corresponding to the feature value, the semantic keywords for classifying the feature items, and the rules for standardizing the feature values. Exemplarily, taking the list code 010504001 of Guangdong in 2013 as an example, the corresponding metadata set is shown in Table 7. It should be noted that the above-mentioned feature value standardization rules are included in the dictionary and the digital regular expression name, which will not be repeated here.
[0143] Table 7
[0144]
[0145] Specifically, in one embodiment, the above step S102 specifically includes the following steps:
[0146] Step S21: Determine the feature item identifier corresponding to the feature item name based on the inventory processing requirements.
[0147] Specifically, the feature item is identified by a special symbol of the feature item name, such as “:”, and the word following the special symbol is the feature item name.
[0148] Step S22: extracting feature item names from the historical building project inventory description set based on the feature item identifiers.
[0149] Specifically, for each historical building project inventory description, the feature item name is extracted according to the above-mentioned special symbols.
[0150] Step S23: construct a feature item set based on the extracted feature item names.
[0151] Therefore, by using the feature item identifiers related to the inventory processing requirements to extract the feature items, the accuracy of the feature item extraction results is improved.
[0152] Specifically, by performing operations such as deduplication and classification on the names of the characteristic items, a characteristic item set is formed. For example, it is assumed that the 20 most frequently appearing characteristic item names are extracted from each list, and the names of the characteristic items are given to the business personnel for screening and verification, and finally the names of the characteristic items are determined to form a characteristic item set. Taking the list code 010504001 in Guangdong in 2013 as an example, the names with ":" are counted from the historical big data as shown in Table 8. After verification by the business personnel, the characteristic item set that affects the quota group price of this list is shown in Table 9.
[0153] Table 8
[0154]
[0155]
[0156] Table 9
[0157] List Sequence Area Code Listing code Feature item name 2013 1710(Guangdong) 010504001 Concrete strength grade Concrete Type Wall Type
[0158] Specifically, in one embodiment, the above step S103 specifically includes the following steps:
[0159] Step S31: extract keywords from the content corresponding to the current feature item.
[0160] Among them, keywords include: text keywords and numeric keywords.
[0161] Step S32: define the category name of the extracted keywords and determine the feature value category.
[0162] Specifically, in an embodiment of the present invention, once the name of the inventory feature item is defined, the NLP tool is used to classify the feature value according to the content of the name, so as to find the name of the most commonly used feature value. The specific steps are: using the NLP tool (the present invention adopts the Stanford similarity calculation method) to classify the content of the feature item name into two types: text type and numeric type; calculating the usage frequency of the text type and numeric type string; handing it over to the business personnel to screen and define the high-frequency text type and numeric type category and name, these names are the category names of the feature value. Screening the feature value name according to the usage frequency of the feature item name content can better eliminate the interference of other words, thereby improving the accuracy of the feature value name, and further helping to improve the accuracy of the inventory description result.
[0163] For example, take the inventory code 010504001 of Guangdong in 2013 as an example: the names of the characteristic items counted from the historical big data include "concrete strength grade", "concrete type", "wall type, and the corresponding characteristic value category names are shown in Table 10.
[0164] Table 10
[0165] List Sequence Area Code Listing code Feature item name Eigenvalue category name 2013 1710(Guangdong) 010504001 Concrete strength grade Concrete strength grade Concrete strength grade Concrete impermeability grade Concrete strength grade Concrete additives Concrete Type Concrete Type Wall Type Wall shape Wall Type Wall Type
[0166] Step S33: establishing a corresponding feature value professional vocabulary dictionary based on the text-type keywords, and establishing a corresponding feature value regular expression based on the numerical keywords.
[0167] Therefore, according to the content of the feature item name, the feature values are classified and the names of the most commonly used feature values are determined, which ensures the consistency between the feature value category and the feature item, making the feature value name more in line with the actual construction project, and establishing a dictionary and regular expression based on the feature value name so that the list can be segmented using the dictionary and regular expression to obtain professional vocabulary that conforms to the actual construction project.
[0168] Specifically, by using NLP tools to classify the contents of the names of the feature items in the list, a dictionary and regular expression for extracting feature values are established from the results. The specific steps are: using NLP tools to classify the contents of the feature item names into text and numeric types, editing the text-type vocabulary into the feature value dictionary, and editing the numeric-type vocabulary into a regular expression.
[0169] For example, take the above-mentioned Guangdong inventory code 010504001 in 2013 as an example: the value expression and the number of statistics of the feature items counted from the historical big data are shown in Table 11. The NLP tool is used to classify the content of the feature item name into two types: text type and numeric type. The text type vocabulary is edited into the dictionary of feature values (vocabulary), and the numeric type is classified into regular expressions (pattern), as shown in Table 12. Among them, the vocabulary sample - concrete additive dictionary example is shown in Table 13, and the pattern sample - concrete impermeability grade example is shown in Table 14.
[0170] Table 11
[0171]
[0172]
[0173]
[0174] Table 12
[0175] Feature item name Eigenvalue category name Pattern Vocabulary Concrete strength grade Concrete strength grade Concrete strength grade Concrete strength grade Concrete impermeability grade Concrete impermeability grade Concrete strength grade Concrete additives Concrete additives Concrete Additives Dictionary Concrete Type Concrete Type Dictionary of concrete names and types Wall Type Wall shape Wall Type Dictionary Wall Type Wall Type Wall Shape Dictionary
[0176] Table 13
[0177]
[0178]
[0179] Table 14
[0180] Impermeability grade Impermeability grade ≥[\s]{0,3}\d\.\dmpa Impermeability grade The anti-seepage grade is not less than [\s]{0,3}\d\.\dmpa Impermeability grade Impermeability level: [\s]{0,3}\d\.\dmpa Impermeability grade The anti-seepage grade is [\s]{0,3}\d\.\dmpa Impermeability grade The impermeability grade is [\s]{0,3}\d\.\dmpa Impermeability grade Impermeability strength grade [\s]{0,3}\d\.\dmpa Impermeability grade Anti-seepage grade [\s]{0,3}\d\.\dmpa
[0181] Therefore, using the statistical methods of machine learning and the similarity tools of NLP vocabulary, the main metadata set is generated, including the feature items of the list, the categories of different values contained in the feature items, the dictionary required to extract the vocabulary corresponding to the feature values, and the regular expressions of the corresponding numeric types of the feature values.
[0182] Specifically, in one embodiment, the above step S105 is implemented by a parser, which is used to extract professional vocabulary from the list description and classify it according to the categories of the feature items and feature values of the list, that is, to structure the list description in the architectural profession, which specifically includes the following steps:
[0183] Step S51: Based on the feature value professional vocabulary dictionary and feature value regular expressions in the metadata set, feature extraction is performed on the engineering content of the construction engineering list to be processed to obtain corresponding professional vocabulary.
[0184] Step S52: Classify the extracted professional vocabulary into characteristic value categories and characteristic items according to the corresponding characteristic value professional vocabulary dictionary or characteristic value regular expression, and generate a structured engineering bill description.
[0185] Specifically, the parser inputs: region code + inventory code + inventory description; outputs: standard feature item name, feature value name, extracted values include normal value and standardized value; algorithm used: metadata set. The code flow of the parser to implement the above process includes: the first step is to read the metadata combination (metadata) corresponding to the inventory, and the second step is to extract the professional vocabulary and professional numbers of the feature items of the inventory.
[0186] a) Read the metadata set corresponding to the manifest (metadata)
[0187] The list metadata set required by the parser includes objects related to parsing basic data, objects related to reading and normalization, and objects related to reading and parsing results, including: list code, list feature items, feature value categories, word association tags, expression specifications, and relationships between dictionaries. The specific metadata set is explained as follows: objects related to reading and parsing basic data: Parsing item: the parsing content and order of each list, Context tag: keywords associated with words, Pattern: regular expressions used to extract fields, Vocabulary: vocabulary tags used to extract fields; objects related to reading and normalization: normalizer_data: standardized units, default_specific_values: standardized special default values, default__comprehensive considerations: standardized "comprehensive considerations" default values, Range_normalizer: standardized Chinese range expressions, for example: no more than =》<=; objects related to reading and parsing results: exclusive_result_valuetype: parsing feature value results that need to be filtered, etc.
[0188] b) Extract professional vocabulary and professional numbers of characteristic items in the list
[0189] The parsing process is the main process of structuring professional vocabulary and professional numbers according to the feature items and feature value categories of an input list data. The main procedures include Pre-process, Segmentation, Token Extraction, Mapping to, Category, Normalizing Results, Building Final Results, etc. For example, Input: area code Dbid, list code itemcode, list name spec_name, and list description spec; Pre-process: In the pre-processing process, special characters are unified and unnecessary words and content are deleted. This process performs two types of cleaning, one before segmentation and the other after segmentation; processing before branching, including changing the list name and list description to lowercase characters, replacing unnecessary description fields, deleting stopwords, and removing unwanted descriptions; Segmentation: The branching process is to use the line breaks and some "symbols" in the list description as line break separators. The descriptions on the same line generally have the same business semantics; define word associations Build Segment Context List: In the embodiment of the present invention, some special words of "context" are used as tags to represent the feature items of the list description. These context tags are associated with the nine-digit list code. Therefore, each list description needs to define the word association. If the context tag keyword is found in the field, the segment will be marked as "context name". For example:
[0190] spec_name: 'Stone-like anti-slip tiles'
[0191] spec: 1.8 thick imitation stone anti-slip brick, 300mm*600mm, M25 (1:1) cement fine sand mortar for sweeping 2.20 thick M20 (1:2.5) cement sand protective layer 3. Polymer cement-based type II waterproof coating 2.0mm thick, 500 high on the walls around 4.20 thick polymer cement waterproof mortar for leveling 5. Cast-in-place reinforced concrete caisson or platform.
[0192] context tag
[0193] Branch 0 'Imitation Stone Anti-slip Tiles':
[0194] ('Block surface material types, mix ratios and specifications',2,3,'Anti-slip bricks',")
[0195] Branch 0 has a word associated label
[0196] Block surface material types, mix ratios and specifications Related keywords Anti-skid brick branch 1'8 thick imitation stone anti-skid brick, 300mm*600mm, M25 (1:1) cement fine sand mortar for sweeping 2.20 thick M20 (1:2.5) cement sand protective layer'
[0197] ('Types, mix ratios and specifications of block surface materials',4,5,'Anti-slip bricks',")
[0198] ('Type and specification of inlay material',33,34,'Seam sweeping',")
[0199] ('Type, mix ratio and thickness of waterproof layer materials',54,56,'protective layer',")
[0200] Branch 1 has three word-related tags:
[0201] Block surface material types, mix ratios and specifications Related keywords Anti-slip bricks
[0202] Types and specifications of fillet materials Related keywords Sweep seam
[0203] Waterproof layer material type, mix ratio and thickness related keywords protective layer branch 2' polymer cement-based type II waterproof coating road 2.0mm thick, 500' high along the walls around
[0204] ('Waterproof layer material type, mix ratio and thickness',9,10,'Waterproof',")
[0205] Branch 2 has a word-related tag:
[0206] Types, mix ratios and thickness of waterproof layer materials Related keywords Waterproof branch 3'20 thick polymer cement waterproof mortar leveling'
[0207] ('Waterproof layer material type, mix ratio and thickness',8,9,'Waterproof',")
[0208] ('Type, mix ratio and thickness of leveling layer materials',12,13,'Leveling',")
[0209] Branch 3 has two word-related tags:
[0210] Types, mix ratios and thickness of waterproof layer materials Related keywords Waterproof
[0211] Types, mix ratios and thickness of leveling layer materials Related keywords Leveling branch 4 'Cast-in-place reinforced concrete caisson or platform'
[0212] There are no word-related tags.
[0213] c) Token Extraction
[0214] The extraction process is divided into two main processes, pattern extraction and vocabulary extraction.
[0215] Exemplarily, according to the above metadata set, the feature value of each feature item of the list has a regular expression name and a dictionary name, for example: the list code 010401001 is shown in Table 15.
[0216] Table 15
[0217]
[0218]
[0219] Among them, the regular expression can have multiple expressions. For example, "Mortar mix ratio" has the following specific expressions: Mortar mix ratio is\d+:\d+:\d+:\d+, Mortar mix ratio\d+:\d+\.\d+
[0220] Mortar mix ratio (?! 3:7)\d:\d, etc. The specific extraction process is to execute all the characteristic values of the characteristic items in the list with pattern to see if there are numbers that meet the expression. If there are, they are the extracted digital words, for example: 'Name: Cement Mortar Name: Cement Mortar 2:3 Cement Mortar, 1:3' The digital word extracted from the mortar mix ratio is: 1:3.
[0221] The dictionary contains collected professional vocabulary. Table 16 shows some of the "basic type dictionaries". If there are matching words in the list description, they will be brought up and then "standardized vocabulary" will be output, that is, similar words are unified into a string. Thus, by standardizing the vocabulary, the structured list description is made more accurate, more flexible, and can adapt to the habits of different designers, etc., expanding the scope of application.
[0222] Table 16
[0223]
[0224]
[0225] Finally, the extracted words are structured (Mapping to Category). This process is to classify the extracted professional words into the feature items and feature value categories of the list. The principle is that the words and regular expressions brought by the feature items and feature value categories will be classified into the same feature item and feature value category. For example: List description: 1. Brick type, specification, strength grade: concrete solid brick 240×115×53 2. Mortar strength grade: DM dry-mix masonry mortar type 1, strength grade: M7.5. Among them, 240*115*53 is extracted from "Specification X*Y*Z" in Table 15, so it should belong to the brick variety, specification, strength grade feature item, brick specification feature category; concrete solid brick is extracted from the block variety dictionary, so it should belong to the brick variety, specification, strength grade feature item, brick variety feature category. Thus, by structuring the list description for construction professionalism, the consistent expression of the list description is achieved, which can not only be used by intelligent pricing products, but also can be used in any scenario that needs to extract professional knowledge from the text description in the construction field, such as intelligent technology customer service products related to construction projects, data collection of professional knowledge bases in the construction field, etc., with a wide range of applications.
[0226] By executing the above steps, the construction project list data processing method provided by the embodiment of the present invention determines the feature items from the historical construction project list descriptions, and determines the feature categories according to the contents of the feature items to establish corresponding feature value professional vocabulary dictionaries and feature value regular expressions, thereby segmenting the professional vocabulary and expressions that conform to the construction field, and structurally classifying the segmented professional vocabulary and expressions into feature items according to the technical meaning of the project content of the list, and only extracting the key words associated with the list processing requirements, so that the final project list description is more in line with the intelligent technology requirements, which is beneficial to improving the performance of intelligent products for construction projects.
[0227] The embodiment of the present invention also provides a method for calculating the engineering quantity. Figure 3 As shown in FIG. 1 , the engineering quantity pricing method specifically includes the following steps:
[0228] Step S201: Obtain a list of construction projects to be priced and a collection of historical reference construction project lists.
[0229] Among them, each historical reference building project list in the historical reference building project list collection has a corresponding historical quota pricing plan, that is, the quota pricing plan carries pricing information, which can be used to calculate the project price.
[0230] Step S202: Set the list processing requirement to quota selection, and use the construction project list data processing method provided by another embodiment of the present invention to process the construction project list to be priced and the historical reference construction project list respectively to obtain the corresponding description of the construction project list to be priced and the historical reference construction project list description set.
[0231] Specifically, by adopting the construction project list data processing method provided by another embodiment of the present invention, the description of the list of projects to be priced and the set of historical reference list descriptions are obtained. Since the list descriptions are unified structured descriptions and only include professional vocabulary and professional numbers related to the quota selection, they provide a data basis for accurate quota group selection for the project quantity pricing of the list of projects to be priced.
[0232] Step S203: Calculate the matching degree between the to-be-priced bill of materials description and each historical reference bill of materials description in the historical reference bill of materials description set.
[0233] Specifically, by calculating the similarity between the character strings in the description of the project list to be priced and the description of each historical reference project list, since all project list descriptions only include professional vocabulary and professional numbers related to the quota selection, and are a unified structured description method, that is, professional vocabulary and professional numbers are classified according to feature items and feature value categories, the accuracy of the similarity results is improved.
[0234] Step S204: Obtain the historical quota group corresponding to the historical reference project list description with the highest matching degree.
[0235] Specifically, the higher the degree of match between the description of the to-be-priced construction project list and the description of the historical reference project list, the smaller the difference in the selection of the quota groups of the two. Therefore, by using the historical quota group corresponding to the historical reference project list description with the highest degree of match as the quota group of the to-be-priced construction project list for project quantity pricing, the accuracy of the project quantity pricing results can be improved.
[0236] Step S205: Calculate the quantities of the construction project list to be priced based on the historical quota group.
[0237] By executing the above steps, the method for costing of project quantities provided by an embodiment of the present invention uses a description method provided by another embodiment of the present invention to perform a structured description on the list of construction projects to be priced and the historical reference list of construction projects. The obtained list description only includes content related to the quota, and the list description is vocabulary and expressions of the construction profession. Matching the quota group with this list description can effectively improve the accuracy of the matching result, thereby improving the accuracy of the costing of project quantities.
[0238] The embodiment of the present invention also provides a construction project list data processing device, such as Figure 4 As shown, the construction project list data processing device includes:
[0239] The first acquisition module 101 is used to acquire a historical building project inventory description set and inventory processing requirements. For details, please refer to the relevant description of step S101 in the above method embodiment, which will not be repeated here.
[0240] The first processing module 102 is used to extract feature items from the historical building project inventory description set based on the inventory processing requirements to obtain a feature item set. For details, please refer to the relevant description of step S102 in the above method embodiment, which will not be repeated here.
[0241] The second processing module 103 is used to determine the feature value category based on the keyword corresponding to each feature item in the feature item set, and establish a corresponding feature value professional vocabulary dictionary and feature value regular expression, which are used to extract professional vocabulary corresponding to the feature value category. For details, please refer to the relevant description of step S103 in the above method embodiment, which will not be repeated here.
[0242] The third processing module 104 is used to construct a metadata set based on the feature item set and the feature value category of each feature item and its corresponding feature value professional vocabulary dictionary and feature value regular expression. For details, please refer to the relevant description of step S104 in the above method embodiment, which will not be repeated here.
[0243] The fourth processing module 105 is used to process the engineering content of the pending construction engineering list based on the metadata set to obtain a structured engineering list description. For details, please refer to the relevant description of step S105 in the above method embodiment, which will not be repeated here.
[0244] For further description of the construction project list data processing device provided in an embodiment of the present invention, please refer to the relevant description of the above-mentioned construction project list data processing method embodiment. The specific implementation processes of the two are similar and will not be repeated here.
[0245] Through the collaborative cooperation of the above-mentioned components, the construction project list data processing device provided by the embodiment of the present invention can segment out professional vocabulary and expressions that conform to the construction field by determining feature items from the description of the historical construction project list, and determining the feature category according to the content of the feature item to establish a corresponding feature value professional vocabulary dictionary and feature value regular expression, and then structuredly classify the segmented professional vocabulary and expressions into feature items according to the technical meaning of the project content of the list, and only extract key words associated with the list processing requirements, so that the final project list description is more in line with the needs of intelligent technology, which is beneficial to improving the performance of intelligent products for construction projects.
[0246] The embodiment of the present invention also provides a device for calculating the engineering quantity. Figure 5 As shown, the engineering quantity pricing device includes:
[0247] The second acquisition module 201 is used to acquire a list of construction projects to be priced and a set of historical reference construction project lists. For details, please refer to the relevant description of step S201 in the above method embodiment, which will not be repeated here.
[0248] The fifth processing module 202 is used to set the list processing requirement to be a quota selection, and use the construction project list data processing device provided by another embodiment of the present invention to process the construction project list to be priced and the historical reference construction project list respectively, and obtain the corresponding description of the construction project list to be priced and the description set of the historical reference construction project list. For details, please refer to the relevant description of step S202 in the above method embodiment, which will not be repeated here.
[0249] The sixth processing module 203 is used to calculate the matching degree between the to-be-priced engineering list description and each historical reference engineering list description in the historical reference engineering list description set. For details, please refer to the relevant description of step S203 in the above method embodiment, which will not be repeated here.
[0250] The seventh processing module 204 is used to obtain the historical quota group corresponding to the historical reference engineering list description with the highest matching degree. For details, please refer to the relevant description of step S204 in the above method embodiment, which will not be repeated here.
[0251] The eighth processing module 205 is used to perform engineering quantity pricing on the list of construction works to be priced based on the historical quota group. For details, please refer to the relevant description of step S205 in the above method embodiment, which will not be repeated here.
[0252] For further description of the engineering quantity pricing device provided in the embodiment of the present invention, please refer to the relevant description of the above-mentioned engineering quantity pricing method embodiment. The specific implementation processes of the two are similar and will not be repeated here.
[0253] Through the coordinated cooperation of the above-mentioned components, the engineering quantity pricing device provided by the embodiment of the present invention uses the description method provided by another embodiment of the present invention to perform structured description on the construction project list to be priced and the historical reference construction project list. The obtained project list description only includes content related to the quota, and the list description is the vocabulary and expression of the construction profession. Matching the quota group with this project list description can effectively improve the accuracy of the matching result, thereby improving the accuracy of the engineering quantity pricing result.
[0254] The embodiment of the present invention further provides an electronic device, such as Figure 6 As shown, the electronic device may include a processor 901 and a memory 902, wherein the processor 901 and the memory 902 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.
[0255] The processor 901 may be a central processing unit (CPU). The processor 901 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.
[0256] The memory 902 is a non-transitory computer-readable storage medium that can be used to store non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions / modules corresponding to the method in the embodiment of the present invention. The processor 901 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions and modules stored in the memory 902, that is, implementing the above method.
[0257] The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required by at least one function; the data storage area may store data created by the processor 901, etc. In addition, the memory 902 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 902 may optionally include a memory remotely arranged relative to the processor 901, and these remote memories may be connected to the processor 901 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0258] One or more modules are stored in the memory 902, and when executed by the processor 901, the above method is performed.
[0259] The specific details of the above electronic device can be understood by referring to the corresponding descriptions and effects in the above method embodiments, and will not be repeated here.
[0260] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.
[0261] 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 above embodiments, a person skilled in the art should understand that the specific implementation modes of the present invention can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.
Claims
1. A method for processing construction project list data, characterized in that: include: Obtain a collection of inventory descriptions of historic buildings and inventory processing requirements; Extracting feature items from the historical building project inventory description set based on the inventory processing requirements to obtain a feature item set; Determine the feature value category based on the keyword corresponding to each feature item in the feature item set, and establish a corresponding feature value professional vocabulary dictionary and feature value regular expression, the feature value professional vocabulary dictionary and feature value regular expression are used to extract professional vocabulary corresponding to the feature value category, each feature item includes multiple feature values, each feature value is collected using a corresponding professional dictionary and definition regular expression, and the feature value category is used to structure the extracted professional vocabulary and digital expressions corresponding to the feature value into the feature items of the corresponding list according to the semantic keywords; Based on the feature item set and the feature value category of each feature item and its corresponding feature value professional vocabulary dictionary and feature value regular expression, a metadata set is formed; The project content of the pending construction project list is processed based on the metadata set to obtain a structured project list description.
2. The method according to claim 1, characterized in that The feature item extraction of the historical building project inventory description set based on the inventory processing requirements to obtain a feature item set includes: Determine a feature item identifier corresponding to the feature item name based on the list processing requirement; Extracting feature item names from the historical building engineering inventory description set based on the feature item identifier; The feature item set is constructed based on the extracted feature item names.
3. The method according to claim 1, characterized in that: The method of determining the feature value category based on the keyword corresponding to each feature item in the feature item set and establishing a corresponding feature value professional vocabulary dictionary and feature value regular expression includes: Extract keywords from the content corresponding to the current feature item, the keywords include: text keywords and numeric keywords; Define the category names of the extracted keywords and determine the feature value categories; A corresponding feature value professional vocabulary dictionary is established based on text keywords, and a corresponding feature value regular expression is established based on numeric keywords.
4. The method according to claim 1, characterized in that The process of processing the engineering content of the pending construction engineering list based on the metadata set to obtain a structured engineering list description includes: Based on the feature value professional vocabulary dictionary and feature value regular expression in the metadata set, feature extraction is performed on the engineering content of the pending construction engineering list to obtain corresponding professional vocabulary; The extracted professional vocabulary is classified into characteristic value categories and characteristic items according to its corresponding characteristic value professional vocabulary dictionary or characteristic value regular expression to generate a structured engineering list description.
5. The method according to claim 1, characterized in that Also includes: Obtaining a list of several historic building works; Based on the area code, list guide version and list code corresponding to the list of building projects to be processed, the description of the list of historical building projects is screened to determine the set of descriptions of the list of historical building projects.
6. A method for calculating the engineering quantity, characterized in that: include: Obtain a list of construction works to be priced and a collection of historical reference construction work lists; The list processing requirement is set to quota selection, and the method described in any one of claims 1 to 5 is used to process the to-be-priced construction project list and the historical reference construction project list, respectively, to obtain a corresponding to-be-priced construction project list description and a historical reference construction project list description set; Calculating the matching degree between the to-be-priced engineering list description and each historical reference engineering list description in the historical reference engineering list description set respectively; Obtain the historical quota group corresponding to the historical reference engineering list description with the highest matching degree; The construction work list to be priced is priced based on the historical quota group.
7. A construction project list data processing device, characterized in that: include: The first acquisition module is used to obtain a collection of historical building project inventory descriptions and inventory processing requirements; A first processing module is used to extract feature items from the historical building project inventory description set based on the inventory processing requirements to obtain a feature item set; The second processing module is used to determine the feature value category based on the keyword corresponding to each feature item in the feature item set, and establish a corresponding feature value professional vocabulary dictionary and feature value regular expression, wherein the feature value professional vocabulary dictionary and feature value regular expression are used to extract professional vocabulary corresponding to the feature value category, each feature item includes multiple feature values, each feature value is collected using a corresponding professional dictionary and definition regular expression, and the feature value category is used to structure the extracted professional vocabulary and digital expressions corresponding to the feature value into the feature items of the corresponding list according to the semantic keywords; A third processing module, configured to construct a metadata set based on the feature item set and the feature value category of each feature item and its corresponding feature value professional vocabulary dictionary and feature value regular expression; The fourth processing module is used to process the project content of the pending construction project list based on the metadata set to obtain a structured project list description.
8. A device for calculating the engineering quantity, characterized in that: include: The second acquisition module is used to obtain a list of construction projects to be priced and a collection of historical reference construction project lists; A fifth processing module, for setting the list processing requirement to quota selection, and using the construction project list data processing device as claimed in claim 7, respectively processing the to-be-priced construction project list and the historical reference construction project list to obtain a corresponding to-be-priced construction project list description and a historical reference construction project list description set; A sixth processing module, used to respectively calculate the matching degree between the to-be-priced engineering list description and each historical reference engineering list description in the historical reference engineering list description set; The seventh processing module is used to obtain the historical quota group corresponding to the historical reference engineering list description with the highest matching degree; The eighth processing module is used to perform engineering quantity pricing on the list of construction works to be priced based on the historical quota group.
9. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which, when executed by a processor, implement the method according to any one of claims 1 to 6.
10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor implements the method as claimed in any one of claims 1 to 6 by executing the computer instructions.
