A bid evaluation method and system based on natural semantic processing

Abstract

The present application relates to a bid evaluation method and system based on natural semantic processing, the method comprising obtaining key words and the association relationship between key words in a bid document, and using the obtained association relationship between key words and key words to form a first spatial grid relationship spectrum; obtaining content associated with the key words in a bid document that matches the key words according to the key words; using the content to adjust the position of the key words in the first spatial grid relationship spectrum to obtain a second spatial grid relationship spectrum; comparing the second spatial grid relationship spectra from two bid documents, and giving a difference degree according to the comparison result. The bid evaluation method and system based on natural semantic processing disclosed in the present application describe the technical specifications in the bid document by a unified processing method, and then discover the differences between different technical specifications by horizontal comparison, so as to improve the standardization and automation of the bid evaluation process.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a bid evaluation method and system based on natural semantic processing. Background Art

[0002] Bid evaluation is the evaluation of the content of the bid document to determine the final winner. Bid documents can be divided into commercial bids, economic bids and technical bids. For commercial bids and economic bids, there are relatively mature automated bid evaluation solutions. For example, the relevant content involved in the commercial bid can be uploaded to the system through the network and then automatically identified and processed by the system. The economic bid can collect information according to the established form and then conduct horizontal comparison.

[0003] Technical bids are mainly used to evaluate and judge the technical strength of bidders. For example, in construction bidding, technical bids are mainly used to examine the performance of construction organization design. This includes the construction organization design of various technical measures taken to complete the project specified in the bidding documents, in order to demonstrate the company's construction organization ability, technical management ability, construction plan and process level.

[0004] The content of the technical bid in the bid varies greatly due to different fields, and the content involved is quite broad. It is difficult to determine a consistent evaluation method. At present, the method of evaluation by experienced personnel is still mainly adopted. However, in large-scale projects, the evaluation by experienced personnel has problems such as involving a large number of people, long reading and comprehension time, differences in manual understanding, differences in industry experience, and other human factors, making it difficult to form a unified evaluation standard. Summary of the invention

[0005] The present application provides a bid evaluation method and system based on natural semantic processing, which describes the technical specifications in the bid documents in a unified manner, and then discovers the differences between different technical specifications through horizontal comparison, so as to improve the standardization and automation of the bid evaluation process.

[0006] The above-mentioned purpose of the present application is achieved through the following technical solutions:

[0007] In a first aspect, the present application provides a bid evaluation method based on natural semantic processing, comprising:

[0008] Obtaining key words and the association relationship between key words in the bidding document, and using the obtained key words and the association relationship between key words to form a first spatial grid relationship spectrum, where the key words come from the bidding document;

[0009] Acquire content associated with the key words in the bidding documents matching the key words according to the key words;

[0010] Using the content to adjust the position of the key words in the first space grid relationship spectrum to obtain a second space grid relationship spectrum;

[0011] The second space grid relationship spectra from the two bid documents are compared, and the difference degree is given according to the comparison result.

[0012] In a possible implementation manner of the first aspect, obtaining key words in the bid document includes:

[0013] Divide the tender document into multiple sentence units according to punctuation marks;

[0014] Identify the subject in each sentence unit to obtain a first subject set;

[0015] Merging repeated first subjects in the first subject set to obtain a second subject set;

[0016] The second subject set is the key words in the bidding document.

[0017] In a possible implementation manner of the first aspect, identifying the subject in each sentence unit includes:

[0018] Divide the sentence units to obtain phrases;

[0019] The phrases are respectively sent to a voting device for voting, and the voting device includes multiple recognition models;

[0020] Determine the meaning of the phrase based on the voting results, and determine the meaning of the sentence unit based on the meaning of the phrase;

[0021] Use the meaning of the sentence unit to determine the subject in the sentence unit.

[0022] In a possible implementation manner of the first aspect, the method further includes:

[0023] The sentence unit is divided multiple times, and the length of the phrases in each division process is the same;

[0024] After the meaning of a phrase is determined, the phrase is taken out of the sentence unit and the remaining parts of the sentence unit are divided.

[0025] In a possible implementation manner of the first aspect, determining the meaning of the sentence unit according to the meaning of the phrase includes:

[0026] Determine the meaning of the phrases, each phrase has at least one meaning;

[0027] Bring each meaning into a sentence unit and explain the sentence unit;

[0028] Input the explanation into a voter for voting, the voter includes multiple recognition models;

[0029] The meanings of the sentence units are ranked according to the voting results. The number of votes a sentence unit receives is negatively correlated with the ranking position of the sentence unit in the sequence.

[0030] The meaning of the first sentence unit in the sequential sequence is determined as the meaning of the sentence unit.

[0031] In a possible implementation manner of the first aspect, using the acquired key words and the association relationship between the key words to form a first spatial grid relationship spectrum includes:

[0032] Assign coordinate points to key words;

[0033] The coordinate points are connected using line segments according to the association relationship to form a first spatial grid relationship spectrum;

[0034] Among them, the association relationship includes inclusion, explanation and extension.

[0035] In a possible implementation manner of the first aspect, using the content to adjust the position of the key words in the first spatial grid relationship spectrum includes:

[0036] Determine the key words that need to be adjusted and record them as the first key words;

[0037] Determine the key words that are directly related to the first key words and record them as the second key words;

[0038] Obtaining content between the first key word and the second key word and extracting the content;

[0039] Calculate the adjustment distance between two key words based on the extracted content;

[0040] The position of the second key word in the first spatial grid relationship spectrum is adjusted using the adjustment distance.

[0041] In a second aspect, the present application provides a bid evaluation device based on natural semantic processing, comprising:

[0042] A first processing unit is used to obtain key words and association relationships between key words in the bidding document, and use the obtained key words and association relationships between key words to form a first spatial grid relationship spectrum, where the key words come from the bidding document;

[0043] A second processing unit is used to obtain content associated with the key words in the tender documents matching the key words according to the key words;

[0044] A position adjustment unit, used to adjust the position of the key words in the first space grid relationship spectrum using the content to obtain a second space grid relationship spectrum;

[0045] The third processing unit is used to compare the second space grid relationship spectra from the two bidding documents and provide a difference degree according to the comparison result.

[0046] In a third aspect, the present application provides a bid evaluation system based on natural semantic processing, the system comprising:

[0047] one or more memories for storing instructions; and

[0048] One or more processors, used to call and run the instructions from the memory to execute the method as described in the first aspect and any possible implementation of the first aspect.

[0049] In a fourth aspect, the present application provides a computer-readable storage medium, the computer-readable storage medium comprising:

[0050] Program, when the program is executed by a processor, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0051] In a fifth aspect, the present application provides a computer program product, comprising program instructions. When the program instructions are executed by a computing device, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0052] In a sixth aspect, the present application provides a chip system, which includes a processor for implementing the functions involved in the above aspects, for example, generating, receiving, sending, or processing the data and / or information involved in the above methods.

[0053] The chip system may be composed of chips, or may include chips and other discrete devices.

[0054] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wired or wireless means, or the processor and the memory can also be coupled on the same device. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic flowchart of the steps of a bid evaluation method provided in this application.

[0056] Figure 2 This is a schematic diagram of the principle of processing a tender document provided by this application.

[0057] Figure 3 This is a schematic diagram provided by the present application for obtaining a second spatial grid relationship spectrum based on a first spatial grid relationship spectrum.

[0058] Figure 4 This is a schematic diagram of adjusting the position of the second key word in the first spatial grid relationship spectrum provided by the present application, and the dotted line corresponds to the position before the adjustment. DETAILED DESCRIPTION

[0059] The technical solution in this application is further described in detail below in conjunction with the accompanying drawings.

[0060] This application discloses a bid evaluation method based on natural semantic processing. Figure 1 In some examples, the bid evaluation method based on natural semantic processing disclosed in this application includes the following steps:

[0061] S101, obtaining key words and the association relationship between the key words in the bidding document, and using the obtained key words and the association relationship between the key words to form a first spatial grid relationship spectrum, where the key words come from the bidding document;

[0062] S102, acquiring content associated with the key words from the bidding documents matching the key words according to the key words;

[0063] S103, using the content to adjust the position of the key words in the first space grid relationship spectrum to obtain a second space grid relationship spectrum;

[0064] S104, comparing the second spatial grid relationship spectra from the two bidding documents, and providing a difference degree according to the comparison result.

[0065] Specifically, in step S101, natural language processing (NLP) technology is first used to extract key words from the tender document based on the key words of the technical bid in the tender document, such as Figure 2 shown.

[0066] For example, the bidding documents contain the following content:

[0067] The new antibody industrialization workshop and its ancillary buildings have a total construction area of ​​61,344.45 square meters. The construction content includes three units: antibody comprehensive workshop, antibody workshop 1, and comprehensive warehouse, except for the pile foundation, gas cylinder shed, hazardous materials warehouse 2, accident emergency pool, guardhouse B, underground garage, outdoor overhead corridor, roads / walls and other outdoor overall.

[0068] The key words obtained at this time include: antibody industrialization workshop, ancillary supporting buildings, antibody comprehensive workshop excluding pile foundation part, antibody workshop 1 excluding pile foundation part, comprehensive warehouse excluding pile foundation part, gas cylinder shed, dangerous goods warehouse 2, accident emergency pool, guard B, underground garage, outdoor overhead corridor, road / wall and other outdoor overall.

[0069] The relationship between the above key terms is: the antibody industrialization workshop includes the antibody comprehensive workshop excluding the pile foundation part and the antibody workshop 1 excluding the pile foundation part; the ancillary supporting buildings include the comprehensive warehouse excluding the pile foundation part, the gas cylinder shed, the dangerous goods warehouse 2, the accident emergency pool, the guardhouse B, the underground garage, the outdoor overhead corridor, the road / wall and other outdoor overall.

[0070] For another example, the bidding documents include the following contents: excavation and backfilling of all earthworks related to rainwater, sewage and wastewater projects inside and outside the factory, procurement and installation of pipes and valves, masonry of pipe wells and procurement and installation of pipe well covers.

[0071] The key words obtained at this time include: inside and outside the factory, rainwater project, sewage project, wastewater project, earthwork, excavation, backfill, pipeline, valve, procurement, installation, pipe well, masonry, and pipe well cover.

[0072] The relationship between the above key words is:

[0073] The indoor and outdoor areas of the factory include rainwater engineering, sewage engineering, and wastewater engineering;

[0074] Stormwater engineering includes earthwork, pipelines, valves, pipe wells, and pipe well covers; sewage engineering includes earthwork, pipelines, valves, pipe wells, and pipe well covers; wastewater engineering includes earthwork, pipelines, valves, pipe wells, and pipe well covers;

[0075] Earthwork includes excavation and backfilling;

[0076] Pipeline includes procurement and installation;

[0077] Valves include procurement and installation;

[0078] Tube wells include masonry;

[0079] Manhole covers include procurement and installation.

[0080] Then, the association relationship between the key words is determined through association analysis. For example, a semantic similarity calculation method is used to determine the association relationship.

[0081] It should be noted here that the bidding documents set out requirements for each node of a project, and the bid needs to be made according to the content given in the bidding documents. Therefore, to some extent, the bidding documents can be regarded as an index of the tender documents. However, different tender documents will display different contents according to their own technical advantages and other aspects.

[0082] In the conventional personnel review method, it is necessary to rely on the reviewers' experience and understanding of the tender documents. Due to the different experiences and understandings of each reviewer, it is difficult to evaluate the technical specifications in the tender documents based on the same standards.

[0083] In order to solve this problem, the present application uses a method of extracting the key words and the association relationship between the key words from the tender documents and then processing them. In step S102, the content associated with the key words is obtained from the tender documents that match the key words according to the key words. These contents are further explanations of the key words. Different tender documents have different explanations, which is the content that the reviewers need to focus on.

[0084] In step S103, by using the associated content extracted from the bid document based on the key words, the position of the key words in the first space grid relationship spectrum is adjusted to obtain a second space grid relationship spectrum, where the second space grid relationship spectrum is based on the evolution of the first space grid relationship spectrum, such as Figure 3 As shown, this evolution method can display different contents in different bidding documents.

[0085] Step S104 , comparing the second spatial grid relationship spectra from the two tender documents, and outputting the difference between the two tender documents according to the comparison result.

[0086] In this way, different bidding documents can be compared horizontally, and the end point is to examine the differences among different bidding documents. In this way, the special content or difference content in the bidding documents can be identified, and then the reviewers can focus on discussing these special contents (difference content).

[0087] It should be understood that if each reviewer needs to read, understand and analyze the tender documents in their entirety, the reviewers will need to read and understand the tender documents in depth, and it will also be necessary to ensure that each reviewer's in-depth reading and understanding is consistent.

[0088] Obviously, this approach is extremely difficult to implement, or even impossible to implement, especially when there are strict time limits for bid evaluation.

[0089] By using the bid evaluation method provided in the present application, the special contents in the bid documents can be taken out from the bid documents, so as to divide the contents in the bid documents into common contents and special contents. In this embodiment, the special contents specifically refer to the different contents compared with other bid documents.

[0090] For reviewers, they can directly evaluate the special content. The advantage of this method is that it reduces the time required for each reviewer to read, understand and analyze the entire tender document, allowing each reviewer to focus more on the differences in the tender documents.

[0091] Of course, in actual applications, there will still be the problem that the differences between the two bidding documents are not obvious, that is, the difference evaluation of the two bidding documents cannot be obtained through the comparison of technical bids, so it is necessary to return to compare the commercial bid and economic bid information in the two bidding documents.

[0092] The technical solution in this application needs to solve two core problems: determining key words and calculating the difference. The specific method for obtaining key words is as follows:

[0093] S201, dividing the tender document into a plurality of sentence units according to punctuation marks;

[0094] S202, identifying the subject in each sentence unit to obtain a first subject set;

[0095] S203, merging repeated first subjects in the first subject set to obtain a second subject set;

[0096] The second subject set is the key words in the bidding document.

[0097] In step S201 to step S203, the bid document will first be divided into multiple sentence units according to punctuation marks. The punctuation marks here include commas, colons, periods, semicolons, etc., or can be described as any symbols that separate the text in the bid document, which are within the scope of punctuation marks referred to in this application.

[0098] Then, the subject in each sentence unit is identified to obtain a first subject set, and then the repeated first subjects in the first subject set are merged to obtain a second subject set. The second subject set obtained at this time is the key words in the bid document.

[0099] The first subject set includes multiple first subjects, which are the subjects in each sentence unit. For the subject in a sentence unit, there may be one explanation or multiple explanations, which increases the number of first subjects. Therefore, it is necessary to merge the repeated first subjects.

[0100] The specific way to identify the subject in each sentence unit is as follows:

[0101] S301, dividing the sentence units to obtain phrases;

[0102] S302, sending the phrases to a voting device for voting, the voting device including a plurality of recognition models;

[0103] S303, determining the meaning of the phrase according to the voting result, and determining the meaning of the sentence unit according to the meaning of the phrase;

[0104] S304, using the meaning of the sentence unit to determine the subject in the sentence unit.

[0105] In step S301 to step S304, the sentence units are first divided to obtain phrases, and then the meaning of the phrases is obtained by voting. It should be noted here that the machine's understanding of phrases is based on a built-in recognition model, and different recognition models have different advantages. In this application, multiple recognition models are used to understand the meaning of phrases, and then voting is used to determine the meaning of the phrases. The specific process is to set a ratio according to the recognition model, for example 0.7. When the recognition models exceeding a ratio give the same result, the meaning is determined as the meaning of the phrase.

[0106] Then, the meaning of the sentence unit is determined according to the meaning of the phrase. The further description of this part is that after the meaning of the phrase is determined, the various components of the sentence unit can be determined. At this time, as long as the subject in the sentence unit is known, the first subject can be directly determined.

[0107] That is, in this method, it is not necessary to completely determine the meaning of the sentence unit, but only to determine the meaning of part of the sentence unit, and the meaning of this part of the sentence unit can support the subject of the sentence unit, which is also within the scope of the technical solution of this application.

[0108] In some possible implementations, the sentence unit is divided multiple times, and the length of the phrases in each division process is the same. At the same time, when the meaning of a phrase is determined, the phrase needs to be taken out of the sentence unit and the remaining part of the sentence unit is continued to be divided.

[0109] In some examples, the specific way to determine the meaning of a sentence unit according to the meaning of a phrase is as follows:

[0110] S401, determining the meaning of a phrase, each phrase having at least one meaning;

[0111] S402, bringing each meaning into a sentence unit and interpreting the sentence unit;

[0112] S403, inputting the explanation into a voting device for voting, the voting device including a plurality of recognition models;

[0113] S404, sorting the meanings of the sentence units according to the voting results, wherein the number of votes obtained by the sentence units is negatively correlated with the sorting position of the sentence units in the sequence;

[0114] S405, determining the meaning of the first sentence unit in the sequential sequence as the meaning of the sentence unit.

[0115] In step S401 to step S405, the meaning of the sentence unit is also determined by voting. The specific method is to use the recognition model to judge the interpretation of the sentence unit, and then determine the meaning of the sentence unit according to the judgment result.

[0116] In some examples, the specific method of using the acquired key words and the association relationship between the key words to form the first spatial grid relationship spectrum is:

[0117] S501, assigning coordinate points to key words;

[0118] S502, connecting the coordinate points using line segments according to the association relationship to form a first spatial grid relationship spectrum;

[0119] Among them, the association relationship includes inclusion, explanation and extension.

[0120] In step S501 and step S502, the key words are converted into coordinate points, and then the association relationships are converted into line segments. These points and line segments constitute the first spatial grid relationship spectrum. Figure 3 , the position of the coordinate point is fixed, and both ends of the line segment are connected to a coordinate point.

[0121] The first spatial grid relationship spectrum obtained at this time can only reflect the basic composition of the tender document, but cannot reflect the differences between different tender documents. For example, if the key words and associations extracted from multiple tender documents have extremely high similarity, then the first spatial grid relationship spectrum obtained at this time will also have extremely high similarity.

[0122] At this time, when comparing the two first spatial grid relationship spectra, there is almost no difference, or there is a difference in the local space as much as possible, and this difference is difficult to reflect the difference between the bid documents.

[0123] In this application, the method of using content to adjust the position of key words in the first space grid relationship spectrum is as follows:

[0124] S601, determining a key word that needs to be adjusted and recording it as the first key word;

[0125] S602, determining a key word that is directly associated with the first key word and recording it as a second key word;

[0126] S603, obtaining content between the first key word and the second key word and extracting the content;

[0127] S604, calculating the adjustment distance between two key words according to the extracted content;

[0128] S605: Use the adjustment distance to adjust the position of the second key word in the first spatial grid relationship spectrum.

[0129] In steps S601 and S602, please refer to Figure 4 , the first keyword and the keyword that is directly related to the first keyword will be selected first. The specific selection method is determined according to the superior-subordinate relationship. That is, when selecting the first keyword, the keyword with the highest level will be selected as the first keyword.

[0130] In some possible implementations, the first key word will be selected based on the bidding documents, and the records in the bidding documents can show the hierarchical relationship between the key words.

[0131] For example, the bidding documents contain the following content: During the excavation of permanent works and construction slopes, building foundation pits, underground caverns, etc., necessary construction safety monitoring instruments should be installed according to the needs of construction safety and (or) the instructions of engineers, necessary construction safety monitoring should be carried out in a timely manner, and the safety monitoring results should be submitted to engineers regularly to prevent settlement, deformation or other damage that may cause safety accidents or affect the normal construction progress.

[0132] The first key words obtained at this time include permanent works, construction slopes, building foundation pits, underground caverns, installation, construction safety monitoring instruments, construction safety monitoring, safety monitoring results, submission, safety accidents, settlement, deformation, and other damages.

[0133] The hierarchical relationship of the above first key words is:

[0134] Permanent works, construction slopes, building foundation pits, and underground caverns are on the same level;

[0135] Permanent works, construction slopes, building foundation pits, and underground caverns include construction safety monitoring; construction safety monitoring includes construction safety monitoring instruments, safety monitoring results, safety accidents or impacts; construction safety monitoring instruments include installation; safety monitoring results include submission; safety accidents or impacts include settlement, deformation, and other damages.

[0136] In some possible implementations, among the key words in the first spatial grid relationship spectrum, only the key words at the end (edge) do not need to be adjusted as the first key words, but will still be regarded as the second key words.

[0137] After obtaining the first key word and the second key word, the content between the first key word and the second key word is obtained and extracted. Here, the method of obtaining the content is as follows:

[0138] Acquire sentence units that are associated with the first key word and the second key word, where the sentence unit refers to the subject in the descriptive content of the first key word and the second key word in the bidding document, and then use these subjects to create a directional spatial grid, in which adjacent subjects in the spatial grid have a direct connection relationship.

[0139] At the same time, different directions are assigned to the attributes of the subject, that is, the spatial grid is composed of multiple direction vectors of the same length. The starting point of this spatial grid is the first key word, and the end point is the second key word. After obtaining the spatial grid, it is necessary to synthesize this spatial grid into a vector, the starting point of the vector is the first key word, and the end point is the second key word.

[0140] Each vector has a direction and a length. Different vectors may have the same or different directions, but the same length. This application uses a classification method to determine the direction of the vector. Each category is assigned a direction, and different categories have different directions.

[0141] Generally speaking, the category is assigned at the initial stage and is a fixed parameter. For unknown categories, a random direction is assigned, and the direction at this time is different from any existing direction.

[0142] At this time, the position of the first key word is fixed, and then the position of the second key word is adjusted according to the obtained vector.

[0143] From the above description, we can see that as the position of the second key word (which needs to be re-determined during each adjustment process) is adjusted, the shape of the second spatial grid relationship spectrum begins to evolve in different directions.

[0144] Based on the above description, a method of comparing the second spatial grid relationship spectra from two bidding documents and giving the difference degree is further described.

[0145] For two second space grid relationship spectra, the same key words will be marked first, and then the offset vector will be obtained through the coordinate difference. Finally, the offset vector will be classified and counted. The classification and statistics method is to first classify the offset vector according to the direction, and then calculate the synthesized length of the offset vector of each category.

[0146] At this time, one or several offset vectors will be obtained. Through the length of the offset vector, the difference between the two bidding documents can be obtained. Through the offset vector and the key words related to the offset vector, the difference content in the bidding document can be quickly determined, and then these difference contents can be evaluated.

[0147] The present application also provides a bid evaluation device based on natural semantic processing, comprising:

[0148] A first processing unit is used to obtain key words and association relationships between key words in the bidding document, and use the obtained key words and association relationships between key words to form a first spatial grid relationship spectrum, where the key words come from the bidding document;

[0149] A second processing unit is used to obtain content associated with the key words in the tender documents matching the key words according to the key words;

[0150] A position adjustment unit, used to adjust the position of the key words in the first space grid relationship spectrum using the content to obtain a second space grid relationship spectrum;

[0151] The third processing unit is used to compare the second space grid relationship spectra from the two bidding documents and provide a difference degree according to the comparison result.

[0152] Furthermore, it also includes:

[0153] The first segmentation unit is used to segment the tender document into a plurality of sentence units according to punctuation marks;

[0154] A sentence recognition unit, used for recognizing the subject in each sentence unit to obtain a first subject set;

[0155] A merging processing unit, used for merging repeated first subjects in the first subject set to obtain a second subject set;

[0156] The second subject set is the key words in the bidding document.

[0157] Furthermore, it also includes:

[0158] The second segmentation unit is used to divide the sentence units into phrases;

[0159] A first voting unit, used to send the phrases to a voting device for voting, wherein the voting device includes a plurality of recognition models;

[0160] A first determination unit, used to determine the meaning of the phrase according to the voting result, and determine the meaning of the sentence unit according to the meaning of the phrase;

[0161] The second determining unit is used to determine the subject in the sentence unit using the meaning of the sentence unit.

[0162] Furthermore, it also includes:

[0163] The third segmentation unit is used to divide the sentence unit multiple times, and the length of the phrases in each division process is the same;

[0164] The phrase processing unit is used to take a phrase out of the sentence unit after the meaning of the phrase is determined, and continue to divide the remaining part of the sentence unit.

[0165] Furthermore, it also includes:

[0166] a third determining unit, for determining the meaning of a phrase, each phrase having at least one meaning;

[0167] Interpretation unit, used to bring each meaning into the sentence unit and interpret the sentence unit;

[0168] A second voting unit, used for inputting the explanation into a voting device for voting, wherein the voting device includes a plurality of recognition models;

[0169] A sorting unit is used to sort the meaning of the sentence units according to the voting results. The number of votes a sentence unit receives is negatively correlated with the sorting position of the sentence unit in the sequence.

[0170] The fourth determining unit is used to determine the meaning of the first sentence unit in the sequential sequence as the meaning of the sentence unit.

[0171] Furthermore, it also includes:

[0172] A fourth processing unit, used for assigning coordinate points to key words;

[0173] A fifth processing unit, configured to connect the coordinate points using line segments according to the association relationship to form a first spatial grid relationship spectrum;

[0174] Among them, the association relationship includes inclusion, explanation and extension.

[0175] Furthermore, it also includes:

[0176] A fifth determining unit, used to determine a key word that needs to be adjusted, recorded as a first key word;

[0177] A sixth determining unit, configured to determine a key word that is directly associated with the first key word and record it as a second key word;

[0178] A sixth processing unit, configured to obtain content between the first key word and the second key word and extract the content;

[0179] A distance calculation unit, used for calculating the adjusted distance between two key words according to the extracted content;

[0180] The distance adjustment unit is used to adjust the position of the second key word in the first spatial grid relationship spectrum using the adjusted distance.

[0181] In one example, the unit in any of the above devices can be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASICs), or, one or more digital signal processors (DSPs), or, one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0182] For another example, when the units in the device can be implemented in the form of a processing element scheduling program, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0183] Various objects such as various messages / information / equipment / network elements / systems / devices / actions / operations / processes / concepts that may appear in this application are named. It can be understood that these specific names do not constitute a limitation on the relevant objects. The names assigned may change with factors such as scenarios, contexts or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from the functions and technical effects embodied / executed in the technical scheme.

[0184] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0185] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0186] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0187] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0188] It should also be understood that in various embodiments of the present application, the first, second, etc. are only used to indicate that multiple objects are different. For example, the first time window and the second time window are only used to indicate different time windows. They should not have any impact on the time window itself, and the first, second, etc. mentioned above should not impose any limitations on the embodiments of the present application.

[0189] It should also be understood that in the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other, and the technical features in different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0190] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a computer-readable storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned computer-readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0191] The present application also provides a bid evaluation system based on natural semantic processing, the system comprising:

[0192] one or more memories for storing instructions; and

[0193] One or more processors are used to call and run the instructions from the memory to execute the method as described above.

[0194] The present application also provides a computer program product, which includes instructions. When the instructions are executed, the bid evaluation system (including terminal devices and network devices) performs operations of the bid evaluation system (including terminal devices and network devices) corresponding to the above method.

[0195] The present application also provides a chip system, which includes a processor for implementing the functions involved in the above content, such as generating, receiving, sending, or processing the data and / or information involved in the above method.

[0196] The chip system may be composed of chips, or may include chips and other discrete devices.

[0197] The processor mentioned in any of the above places can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for executing programs for controlling the above-mentioned feedback information transmission method.

[0198] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wire or wireless means to support the chip system to implement various functions in the above embodiments. Alternatively, the processor and the memory can also be coupled on the same device.

[0199] Optionally, the computer instructions are stored in a memory.

[0200] Optionally, the memory is a storage unit within the chip, such as a register, a cache, etc. The memory can also be a storage unit within the terminal located outside the chip, such as a ROM or other types of static storage devices that can store static information and instructions, RAM, etc.

[0201] It can be understood that the memory in the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0202] The non-volatile memory may be a ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory.

[0203] The volatile memory may be a RAM, which is used as an external cache. There are many different types of RAM, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct memory bus RAM.

[0204] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A bid evaluation method based on natural semantic processing, characterized in that: include: Acquire key words and association relationships between key words in the bidding document, and form a first spatial grid relationship spectrum based on the association relationships between the key words and the key words, wherein the key words come from the bidding document; Acquire content associated with the key words in the bidding documents matching the key words according to the key words; Using the content to adjust the position of the key words in the first space grid relationship spectrum to obtain a second space grid relationship spectrum; The second space grid relationship spectra from the two bid documents are compared, and the difference degree is given according to the comparison result.

2. The bid evaluation method based on natural semantic processing according to claim 1 is characterized in that: The key words in obtaining the tender documents include: Divide the tender document into multiple sentence units according to punctuation marks; Identify the subject in each sentence unit to obtain a first subject set; Merging repeated first subjects in the first subject set to obtain a second subject set; The second subject set is the key words in the bidding document.

3. The bid evaluation method based on natural semantic processing according to claim 2 is characterized in that: Identify the subject in each sentence unit including: Divide the sentence units to obtain phrases; The phrases are respectively sent to a voting device for voting, and the voting device includes multiple recognition models; Determine the meaning of the phrase based on the voting results, and determine the meaning of the sentence unit based on the meaning of the phrase; Use the meaning of the sentence unit to determine the subject in the sentence unit.

4. The bid evaluation method based on natural semantic processing according to claim 3 is characterized in that: Also includes: The sentence unit is divided multiple times, and the length of the phrases in each division process is the same; After the meaning of a phrase is determined, the phrase is taken out of the sentence unit and the remaining parts of the sentence unit are divided.

5. The bid evaluation method based on natural semantic processing according to claim 3 or 4, characterized in that: Determining the meaning of a sentence unit based on the meaning of a phrase includes: Determine the meaning of the phrases, each phrase has at least one meaning; Bring each meaning into a sentence unit and explain the sentence unit; Input the explanation into a voter for voting, the voter includes multiple recognition models; The meanings of the sentence units are ranked according to the voting results. The number of votes a sentence unit receives is negatively correlated with the ranking position of the sentence unit in the sequence. The meaning of the first sentence unit in the sequential sequence is determined as the meaning of the sentence unit.

6. The bid evaluation method based on natural semantic processing according to claim 1 is characterized in that: The first spatial grid relationship spectrum is formed by using the acquired key words and the association relationship between the key words, including: Assign coordinate points to key words; The coordinate points are connected using line segments according to the association relationship to form a first spatial grid relationship spectrum; Among them, the association relationship includes inclusion, explanation and extension.

7. The bid evaluation method based on natural semantic processing according to claim 6 is characterized in that: Using content to adjust the position of key words in the first space grid relationship spectrum includes: Determine the key words that need to be adjusted and record them as the first key words; Determine the key words that are directly related to the first key words and record them as the second key words; Obtaining content between the first key word and the second key word and extracting the content; Calculate the adjustment distance between two key words based on the extracted content; The position of the second key word in the first spatial grid relationship spectrum is adjusted using the adjustment distance.

8. A bid evaluation device based on natural semantic processing, characterized in that: include: A first processing unit is used to obtain key words and association relationships between key words in the bidding document, and use the obtained key words and association relationships between key words to form a first spatial grid relationship spectrum, where the key words come from the bidding document; A second processing unit is used to obtain content associated with the key words in the bid document matching the key words according to the key words; A position adjustment unit, used to adjust the position of the key words in the first space grid relationship spectrum using the content to obtain a second space grid relationship spectrum; The third processing unit is used to compare the second space grid relationship spectra from the two bidding documents and provide a difference degree according to the comparison result.

9. A bid evaluation system based on natural semantic processing, characterized in that: The system comprises: one or more memories for storing instructions; and One or more processors, configured to call and execute the instructions from the memory to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer readable storage medium comprises: The program, when the program is executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images