Bidding and tendering information recommendation method and related equipment

Abstract

The invention discloses a bidding and tendering information recommendation method and related equipment. The method comprises the following steps: identifying key time node information of a target bid invitation announcement so as to construct a project qualification validity time axis under the condition that time nodes cover enrollment cut-off, bid opening time, contract signing and plan completion; based on a pre-stored qualification file matched with the bidding subject, querying a certificate validity period, an annual inspection and certificate change plan and a key personnel certificate state associated with the qualification file; and carrying out time sequence matching according to a project qualification validity time axis and the certificate validity period, determining a qualification coverage rate from an application day to a planned completion day, and prompting the investability of the current marking table. The problem that bidding efficiency is low or fair competition is difficult to realize due to the fact that bidding information is highly dispersed in public resource transaction platforms, industry association websites, enterprise self-built platforms and even media reloading, and bidders are difficult to quickly screen out projects which can be casted and winner in a large amount of information can be solved.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the computer field, and more particularly, to a bidding information recommendation method and related equipment. BACKGROUND

[0002] Bidding originates from the need of public procurement and large-scale engineering or service procurement to achieve value for money and prevent rent-seeking. In the government, state-owned enterprises and large private enterprises, the procurement amount is high, the cycle is long, and many parties are involved. Therefore, the demand, qualification, price and risk must be transparentized through standardized processes. Typical participants include bidders, potential bidders, evaluation experts and regulatory authorities. Typical processes include a series of processes such as announcement release, qualification / reporting, clarification, bidding, bid opening, bid evaluation, bid-winning publicity, and contract signing and performance.

[0003] However, due to the high dispersion of bidding information in various levels of public resource trading platforms, industry association websites, enterprise self-built platforms and even media reprints, it is difficult for bidders to quickly screen projects that can be bid and have a chance of winning from a large amount of information, resulting in low bidding efficiency or difficulty in achieving fair competition. SUMMARY

[0004] A series of simplified concepts are introduced in the summary section, which will be further described in detail in the specific embodiments section. The summary section of the present application does not mean to attempt to limit the key features and necessary technical features of the claimed technical solutions, nor to attempt to determine the protection scope of the claimed technical solutions.

[0005] In order to solve the problem that due to the high dispersion of bidding information in various levels of public resource trading platforms, industry association websites, enterprise self-built platforms and even media reprints, it is difficult for bidders to quickly screen projects that can be bid and have a chance of winning from a large amount of information, resulting in low bidding efficiency or difficulty in achieving fair competition, in a first aspect, the present application provides a bidding information recommendation method, the method comprising:

[0006] Identifying key time node information of a target bidding announcement to construct a project qualification validity time axis in the case that the key time node information covers the bid closing date, the bid opening time, the contract signing and the planned completion date;

[0007] Based on the pre-stored qualification files matched by the bidding subject, querying the certificate validity period, annual inspection and certificate replacement plan, and key personnel certificate status associated with the qualification files;

[0008] According to the time sequence matching of the project qualification validity time axis and the certificate validity period, determining the qualification coverage rate from the bid date to the planned completion date, and prompting the bidability of the current bid.

[0009] Optionally, it further comprises:

[0010] Generate a certificate buffer or personnel replacement suggestion in the case of coverage below a preset threshold.

[0011] Optionally, further comprising:

[0012] Obtain the personnel and certificate pre-occupancy time information of the bidding subject in parallel projects, and construct a cross-project qualification conflict graph in the case that the pre-occupancy time window covers the key nodes of the bid closing, review and performance of the current bid;

[0013] Based on the pre-stored qualification archives matched by the bidding subject, query the on-site commitment, shift plan and labor compliance restrictions associated with the qualification archives;

[0014] According to the timing matching of the conflict graph and the project qualification validity time axis, predict the cross-project qualification available time slot, generate a personnel scheduling or replacement scheme suggestion in the case of insufficient available time slot, and prompt the bidability of the current bid.

[0015] Optionally, further comprising:

[0016] Parse the bill of quantities information of the target tender announcement;

[0017] In the case that the bill of quantities information includes item coding, unit of measurement and work content description, respectively extract the item level to generate an item-weight mapping table, the weight being the proportion of the item amount to the total amount of the list, the item amount being calculated based on quantity and unit price;

[0018] Based on the matched industry benchmark list and historical winning samples, query the item fingerprint information corresponding to the announcement list, the fingerprint containing the typical level depth, common item set and reference interval of each item weight of similar projects;

[0019] According to the alignment result of the item fingerprint information and the announcement list, calculate the structural entropy and weight deviation rate using the item-weight mapping table to determine excessive splitting or abnormal merging, and prompt the bidability of the current bid.

[0020] Optionally, the calculating the structural entropy and weight deviation rate using the item-weight mapping table to determine excessive splitting or abnormal merging, and prompting the bidability of the current bid, comprises:

[0021] In the case that the structural entropy deviates from the preset threshold or the weight deviation rate is abnormally concentrated, generate an uncomparable price or induced low price risk prompt;

[0022] Output the bid allocation and clarification question and answer suggestion of the key item.

[0023] Optionally, further comprising:

[0024] extracting adjacent and hierarchical sequence relations of the sub-items based on the bill of quantities information;

[0025] querying an adjacent transfer probability matrix fingerprint of a similar bill based on an industry sample;

[0026] constructing a transfer matrix according to the bill of quantities information and calculating a KL divergence with the adjacent transfer probability matrix fingerprint to judge an abnormal split in divergence anomaly and to prompt the bidability of the current bid document.

[0027] Optionally, the method further comprises:

[0028] extracting work content descriptions and measurement rules of each sub-item based on the bill of quantities information to generate corresponding semantic vectors;

[0029] calculating semantic similarities between sub-items according to the semantic vectors to form a sub-item similarity matrix;

[0030] calculating range overlap coefficients between sub-items based on the measurement rules of the sub-items;

[0031] determining an overlap degree hot area according to the sub-item similarity matrix and the range overlap coefficients and calculating a repeated measurement index;

[0032] generating a repeated measurement prompt in a case where the repeated measurement index exceeds a threshold value.

[0033] In a second aspect, the present application further provides a bid recommendation device, comprising:

[0034] an identification unit configured to identify key time node information of a target bid announcement to construct a project qualification validity time axis in a case where the key time node information covers a bid submission deadline, a bid opening time, a contract signing and a planned completion;

[0035] a query unit configured to query a certificate validity period, an annual inspection and certificate replacement plan and a key personnel certificate status associated with a pre-stored qualification file matched by a bid subject based on the qualification file;

[0036] a determination unit configured to determine a qualification coverage rate from a bid submission date to a planned completion date according to a time sequence matching between the project qualification validity time axis and the certificate validity period to prompt the bidability of the current bid document.

[0037] In a third aspect, an electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the bid recommendation method of any one of the first aspect when executing the computer program stored in the memory.

[0038] In a fourth aspect, the present application further provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the bidding information recommendation method according to any one of the first aspect.

[0039] To sum up, the bidding information recommendation method provided by the present application can recognize key time node information of a target bidding announcement, so as to construct a project qualification validity time axis in the case that the key time node information covers the registration deadline, the bid opening time, the contract signing and the planned completion; query the certificate validity period, the annual inspection and the certificate replacement plan and the key personnel certificate state associated with the pre-stored qualification file matched by a bidding subject; and determine the qualification coverage rate from the registration day to the planned completion day according to the time sequence matching of the project qualification validity time axis and the certificate validity period, and prompt the bidability of the current bid document. Therefore, for the inconsistency of the effectiveness window of the qualification certificate at different time points of registration, bid opening and performance, the conventional matching only checks whether the current one is effective. The present application can accurately identify the time sequence risks such as qualification invalidation on the bid opening day, annual inspection interruption in the middle of performance and personnel certificate replacement window, which are easily missed in traditional manual review. The qualification coverage rate measures the compliance degree throughout the whole process with a single numerical value, and superimposes rigid checking of key nodes, which can avoid both the misjudgment of high coverage rate but invalidation at key moments and the optimistic deviation of node validity but interruption in the middle. The standardized time axis and interval segments make the prompt results traceable, accountable and reviewable, which significantly reduces the communication cost and compliance disputes. In engineering practice, this determination strategy can also be linked with the recommendation system to preferentially push the bid document with high qualification continuity and safe key nodes, which can improve the bid hit rate and reduce the cost risk caused by invalid registration, flow bidding or performance obstruction.

[0040] The bidding information recommendation method of the present application, other advantages, objects and features of the present application will be embodied in part by the following description, and will be understood by those skilled in the art through research and practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0041] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are intended to further assist in understanding the preferred embodiments, and are not considered limiting of the present description. Moreover, like reference numerals denote like parts throughout the several views in the drawings. In the drawings:

[0042] Figure 1 A bidding information recommendation method flowchart provided by an embodiment of the present application;

[0043] Figure 2 A bidding information recommendation device structure schematic diagram provided by an embodiment of the present application;

[0044] Figure 3 A bidding information recommendation electronic device structure schematic diagram is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0045] The terms "first", "second", "third", "fourth" and the like in the description and claims of the present application and the above drawings (if any) are used to distinguish similar objects, and are not necessarily used to describe a particular sequential or chronological order. It should be understood that the data thus used can be interchanged, where appropriate, so that the embodiments described herein can be carried out in sequences other than those illustrated or described herein. Furthermore, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, processes, methods, systems, products, or devices that include a series of steps or units not necessarily limited to those clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products, or devices. The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all.

[0046] In order to solve the problem that due to the high dispersion of bidding information in various levels of public resource trading platform, industry association website, enterprise self-built platform and even media reprint, it is difficult for the bidder to quickly screen the project that can be bid and has a chance of winning from a large amount of information, resulting in low efficiency of bidding and difficult to achieve fair competition, please refer to Figure 1 A bidding information recommendation method flowchart is provided for the embodiments of the present application, which can specifically include steps S110 to S130.

[0047] S110, identifying key time node information of a target bidding announcement, so as to construct a project qualification validity time axis in the case that the key time node information covers the registration deadline, the opening time, the contract signing and the planned completion.

[0048] S120, based on the pre-stored qualification archives matched by the bidding subject, querying the certificate validity period, annual inspection and certificate replacement plan and key personnel certificate status associated with the qualification archives.

[0049] S130, according to the time sequence matching of the project qualification validity time axis and the certificate validity period, determining the qualification coverage rate from the registration date to the planned completion date, and prompting the bidability of the current bid.

[0050] For example, the project qualification validity timeline can be constructed first. The system parses the time fields such as the start and end of the registration, the tender / opening, the contract signing or the project duration starting point, the planned completion or delivery date, etc. from the target tender announcement. When the announcement does not explicitly plan the completion but provides the duration, the planned completion date is extrapolated from the contract or the bid notice starting point. When there is a clarification or change announcement, the latest valid version covers the original field. Then set the registration date as the interval starting point and the planned completion date as the interval ending point to form the project qualification validity timeline, which is the only time reference for all subsequent calculations and comparisons. The precise timestamps of the four key nodes are recorded for subsequent rigid checking and audit tracing.

[0051] For example, the system retrieves the validity period of the enterprise qualification certificate, production / operation license, relevant system certification, and key personnel certificate based on the pre-stored qualification archives matched by the tender subject. The information such as the validity period, annual inspection time window, expected effective time of license replacement / registration plan, necessary continuing education or on-duty status, etc. is obtained. One or more valid interval segments are generated for each certificate, such as "current certificate valid until X day" or "valid again from Y day after license replacement is completed". If the announcement / regulation does not explicitly accept the period during annual inspection or license replacement, it is considered valid. If not, it is treated as invalid before the license is effective. If the announcement explicitly requires that one person cannot hold multiple positions simultaneously or requires full-time on-site personnel, the effective interval is divided according to the occupation conflict based on the personnel roster / occupation information. Finally, a standardized effective interval set corresponding to each required qualification is obtained, which has been processed for changes and occupation constraints.

[0052] For example, the system first constructs the required qualification list based on the explicit hard requirements in the tender document, such as enterprise qualification level, specific license / certification, project manager registration category and level, etc. Then, the effective interval of each item in the list is projected onto the time axis obtained in the first step, and the simultaneous validity is determined within the entire project interval. If all required qualifications are in an effective state on a certain day or finer granularity time slice, it is marked as qualification valid. The system calculates the qualification coverage rate in the project time axis. The four key nodes of registration deadline, opening, contract signing, and planned completion are checked one by one to determine whether they are in the qualification valid state. Based on the configured threshold and key node checking results, the tenderability prompt is output, such as tenderable / boundary tenderable / untenderable. The qualification name, gap start and end time, and corresponding evidence field that cause discontinuity are included in the prompt to ensure explainability and auditability.

[0053] It can be understood that, since the discrete nodes of the bidding document, such as the registration deadline, the bidding time, the contract signing and the planned completion, are uniformly mapped to the continuous time axis from the registration date to the planned completion date, the time axis is taken as the judgment interval; then each item of qualification that the bidding subject must have, such as enterprise certificate, license, system certification, validity period of key post personnel certificate, annual inspection and certificate replacement plan, on-the-job status and the like, is unified into one or more effective intervals; at any time, only when all the necessary qualifications are valid at the same time, the subject has the implementation qualification at the time, and thus the qualification validity time within the project period can be obtained by the intersection of the set of effective intervals; the proportion of the effective time obtained by the intersection in the project time axis is taken as the qualification coverage rate, and the four key nodes of the registration deadline, the bidding, the contract signing and the planned completion are taken as the rigid checking points for superimposed judgment, so as to objectively quantify the whole-process compliance and the key-point compliance, thereby prompting the bidability of the current bid document.

[0054] In summary, the bidding and tendering information recommendation method provided by the embodiments of the present application identifies the key time node information of the target bidding announcement, constructs a project qualification validity time axis in the case that the key time node information covers the registration deadline, the bidding time, the contract signing and the planned completion, queries the certificate validity period, the annual inspection and certificate replacement plan and the key personnel certificate state associated with the pre-stored qualification archives of the bidding subject based on the qualification archives, determines the qualification coverage rate from the registration date to the planned completion date by time sequence matching of the project qualification validity time axis and the certificate validity period, and prompts the bidability of the current bid document. Thus, for the inconsistency of the validity window of the qualification certificate at different time points of registration, bidding and performance, the conventional matching only checks whether the current one is valid, and the present scheme can accurately identify the time sequence risks such as qualification invalidation on the bidding day, annual inspection interruption during performance and personnel certificate replacement window, which are easily missed by traditional manual review. The qualification coverage rate measures the whole-process compliance degree with a single numerical value, and the key node rigid checking avoids both the false judgment of high coverage rate but invalidation at the key moment and the optimistic deviation of node validity but interruption in the middle section. The standardized time axis and interval segments make the prompt result traceable, accountable and reviewable, significantly reducing the communication cost and compliance disputes. In engineering practice, this determination strategy can also be linked with the recommendation system to preferentially push the bid documents with high qualification continuity and safe key nodes, which empirically improves the bid hit rate and reduces the cost risk caused by invalid registration, flow bidding or performance obstruction.

[0055] In some examples, further comprising:

[0056] Generating a certificate renewal buffer or personnel replacement suggestion in the case that the coverage rate is lower than a preset threshold.

[0057] Exemplarily, in the case that the qualification coverage is lower than the preset threshold, the specific qualification gap start and end dates are located based on the project qualification validity timeline, and the responsible qualification or responsible post personnel causing the gap are identified; after the identification is completed, the earliest effective date and the recommended safety buffer days are estimated in combination with the annual inspection or replacement handling time length of the responsible qualification, the required material completeness, the administrative suspension calendar, and the reservation availability, if the earliest effective date can meet the safety buffer before the gap start date, a certificate renewal buffer suggestion is generated, and the above certificate renewal buffer suggestion at least includes the suggested latest date to start handling, the handling path and material list, the estimated effective date and the recommended buffer days, the impact assessment on key nodes such as the registration deadline and the bid opening; in the case that the earliest effective date cannot cover the gap or the key nodes, the personnel replacement path is entered, the personnel that match the certificate type and level, have no project occupation conflicts during the gap period, and meet the on-site or full-time constraints are selected from the candidate personnel library, and a single-person solution that can cover the entire gap period is preferentially selected; in the case that there is no single-person coverage, a minimum replacement combination solution is generated, the on-duty start and end time of each replacement personnel, the handover time point and handover list, the labor compliance upper limit, and the shift connection rules are specified, and the qualification coverage and key node validity under the replacement solution are recalculated and compared and displayed synchronously, so as to prompt the user in a structured manner how to restore the coverage to reach the threshold. Thus, by accurately positioning the qualification gap to the responsible qualification or responsible personnel, and converting the certificate renewal and replacement two paths into a calculable problem of whether the earliest effective time covers the gap, an executable and auditable repair solution can be given without changing the bid time nodes; compared with only prompting that it cannot be bid, the risk of losing the certificate or invalidating the key nodes in the middle can be significantly reduced, and the recalculation coverage and key node checking results are used as closed-loop evidence to improve the certainty and compliance of the decision.

[0058] In some examples, further comprising:

[0059] Obtaining the personnel and certificate pre-occupation time information of the bid subject in parallel projects, and constructing a cross-project qualification conflict graph in the case that the pre-occupation time window covers the registration deadline, review, and performance key nodes of the current bid;

[0060] Based on the pre-stored qualification archives matched by the bid subject, the on-site commitment, shift plan, and labor compliance restrictions associated with the qualification archives are queried;

[0061] According to the conflict graph and the project qualification validity timeline, the time sequence is matched, the cross-project qualification available time slots are predicted, the personnel scheduling or replacement solution suggestions are generated in the case that the available time slots are insufficient, and the bidability of the current bid is prompted.

[0062] Exemplarily, certificate pre-occupancy time information of personnel of a bidding subject in parallel projects can be acquired, the pre-occupancy time including on-site start and end, certificate occupation and site residence period of each candidate personnel and the corresponding certificate in other projects; in the case that the pre-occupancy time window covers the key nodes of bidding, evaluation and performance of the current bidding case, a cross-project qualification conflict graph is constructed, the graph taking personnel / certificate, key period of the current bidding case / period of parallel projects as nodes, time overlap / region inaccessibility / site residence conflict as edges and marking severity; based on the pre-stored qualification archives matched by the bidding subject, site residence commitments associated with the qualification archives such as single-project full-time / continuous site residence days / must-on-site period, shift plans such as shift, working hours and shift limit and labor compliance limits such as daily / weekly working hour upper limit, mandatory rest, one certificate one project, etc. are queried; according to the conflict graph and the project qualification validity time axis, time sequence matching is performed, the available time slot set of each key post is calculated = the required time slot of the current bidding case - the intersection of the pre-occupancy time slot of the parallel project and the compliance disabled time slot, and whether the available time slot covers the continuous requirements of the bidding deadline, evaluation and performance is checked; in the case that the available time slot is insufficient, personnel scheduling or substitution scheme suggestions are generated according to the priority of shift, substitution to the minimum combination replacement, the on-duty / handover time point and compliance verification result are specified, and the qualification coverage rate and key node validity are recalculated, and the bidability of the current bidding case is prompted.

[0063] It can be understood that the problem of converting the cross-project personnel certificate occupation and compliance constraints into time interval difference set and conflict graph can quantify whether there is a person who can legally work in the key period before bidding, and give executable scheduling / substitution scheme and recalculation evidence, reduce the risk of no one available at the opening / bid performance node, and improve the certainty and auditability of the bidability judgment.

[0064] In some examples, further comprising:

[0065] Parsing the engineering quantity list information of the target bidding announcement;

[0066] In the case that the engineering quantity list information includes sub-item coding, measurement unit and work content description, the sub-item level is extracted respectively to generate a sub-item-weight mapping table, the weight being the proportion of the sub-item amount to the total amount of the list, the sub-item amount being calculated based on quantity and unit price;

[0067] Based on the matched industry benchmark list and historical winning samples, sub-item fingerprint information corresponding to the announcement list is queried, the fingerprint including typical level depth, common sub-item set and reference interval of each sub-item weight of similar projects;

[0068] According to the alignment result of the sub-item fingerprint information and the announcement list, the structural entropy and weight offset rate are calculated by using the sub-item-weight mapping table to determine the over-splitting or abnormal merging, and the current bidding is prompted.

[0069] It can be understood that the structure information and weight (amount ratio information) of the bill of quantities BoQ can be unified into the comparable price measurement space. First, the announcement list is parsed into a standardized representation of sub-item-quantity-unit price-amount-amount ratio. Then, the sub-item fingerprint of the industry benchmark list and the historical winning sample is taken as a reference. After alignment, two dimensions are judged for abnormalities: one is whether the structural complexity of the list is abnormally increased or abnormally contracted to identify over-splitting or abnormal merging; the other is whether the amount ratio of the key sub-item significantly deviates from the reference interval to identify the risk of incomparable price and induced low price. The dual check of structure and weight can convert whether the price is comparable from experience judgment to quantifiable determination, and accordingly prompt the current bidding.

[0070] Exemplary, the system performs layout recognition and structure recovery on the announcement body and its attachments such as HTML tables, PDF and scanned images, extracts the table header and row item information, including sub-item code, sub-item name, unit of measurement, work content description, quantity, control unit price / total price, chapter level identification, etc.; performs OCR on the scanned images and recovers the field boundaries combined with table line and column alignment, reorganizes the merged cells and line breaks; performs term normalization and entity disambiguation on synonymous items and abbreviations such as "soil removal / garbage removal" or "temporary building / temporary facilities"; data cleaning on duplicate rows, cross-page line breaks, and abnormal characters to ensure consistency and traceability of subsequent hierarchical reconstruction and amount aggregation. Under the premise that the list information is complete, i.e. at least contains sub-item code, unit of measurement, work content description, the list hierarchy tree is reconstructed according to the chapter-department-item-subject rules, and the sub-item amount is calculated line by line: preferentially use the unit price / total price given in the announcement; if not given, derive the reference unit price according to the industry benchmark price library and regional / distance / period context factors, and then multiply it by the quantity to get the sub-item amount; use the proportion of each sub-item amount in the total amount of the list as the weight to form a mapping table of sub-item identification-hierarchy path-quantity-unit price source-amount-weight; label the source and confidence of the missing or estimated unit price, and complete the amount closure check, such as sub-item summation and total amount consistency and negative / abnormal value detection, to provide data basis for subsequent alignment with fingerprints and abnormal explanation. Based on project type, professional category and region, the announcement list is mapped to the industry benchmark list and historical bid samples that have been settled, and the corresponding sub-item fingerprints are retrieved, which at least include: typical hierarchical depth and level distribution of similar projects, common and mandatory sub-item set, reasonable range of key sub-item amount proportion and its dispersion, common split and merge granularity and measurement boundary description, etc.; then align with the hierarchical path, standard terminology and similar text as anchor points to establish one-to-one, one-to-many or many-to-one mapping relationship between the announcement sub-item and the fingerprint sub-item, label the newly added / atypical sub-item that fails to align, and the sub-item that is merged by multiple typical sub-items, and record the alignment evidence such as term matching, similarity and historical co-occurrence, etc. for judgment.After alignment, the system first evaluates the structural complexity of the list according to the amount proportion distribution of the sub-item-weight mapping table, combined with the common level depth and granularity of the fingerprint, to assess the business meaning of the structure entropy: the finer the sub-items and the greater the amount dispersion, the higher the complexity; otherwise, the lower the complexity. When the complexity is significantly higher than the reference interval, it is determined that there may be excessive splitting; when it is significantly lower than the reference interval, it is determined that there may be abnormal merging. Then, for all items aligned to the key sub-items, check whether the amount proportion falls within the fingerprint reference interval, and calculate the overall deviation, i.e. the weight deviation rate, to identify the induced low price space by lowering high proportion items or raising marginal items. By integrating the evidence of structural complexity deviation, key sub-item deviation, and unaligned or merged suspicion, the risk type and evidence list are generated, and targeted suggestions are output, such as suggesting to split a merged item according to the fingerprint granularity, supplementing the measurement boundary, clarifying the temporary listing and measure fee listing criteria, or giving a bid allocation suggestion for abnormal deviation items. Finally, combined with the business strategy of the enterprise, the conclusion is mapped into a actionable prompt with attached impact clauses and clarification question and answer points, facilitating target selection and communication preparation before bidding.

[0071] It can be understood that, based on the above scheme, the two indicators of structural complexity and key sub-item weight deviation will convert the previous experience-based comparability / induced low price into quantifiable conclusions that can be reviewed and audited, significantly reducing subjective differences. Excessive splitting and abnormal merging are often hidden under the format and terminology. This method uses fingerprint alignment and level reconstruction to flatten the criteria, allowing for early risk identification and reducing ineffective investment. The output not only has conclusions, but also evidence chains and operational suggestions, directly serving pre-bid clarification and bid strategy, reducing passive responses. The fingerprint is derived from industry and historical bidding samples, and can be updated automatically with industry changes to ensure wide applicability in different regions and different professions. Coupling the risk determination results with the recommendation system, the system prioritizes well-structured and healthy weight distribution tenders, improving bid hit rate and reducing compliance and performance risks.

[0072] In some examples, the calculating the structure entropy and the weight deviation rate using the sub-item-weight mapping table to determine excessive splitting or abnormal merging, and prompting the bidability of the current tender, includes:

[0073] generating a non-comparable or induced low price risk prompt when the structure entropy deviates from a preset threshold or the weight deviation rate is abnormally concentrated;

[0074] outputting bid allocation and clarification question and answer suggestions for key sub-items.

[0075] It can be understood that,

[0076] For example, the system reads the sub-item-weight mapping table generated by the pre-processing, counts the weight distribution of each layer according to the list hierarchy such as division / sub-item / sub-sub-item, etc., and compares it with the reference interval of the structural entropy of the same type of project; when the structural entropy of the whole or a certain key level is significantly higher than the upper reference interval, it is marked as possible over-splitting, and the largest contribution node is located, such as a cluster of sub-items under the same parent item that appears abnormally large in number and generally small in amount ratio; when the structural entropy is significantly lower than the lower reference interval, it is marked as possible abnormal merging, and the aggregated node suspected of being merged is located, such as a parent item with an abnormally high amount ratio and few sub-items; at the same time, evidence fields such as hierarchical path, line number, weight distribution quantile value and corresponding industry reference file are recorded to ensure the explainability and auditability of subsequent prompts. After the system completes the alignment of terms and levels with the sub-item fingerprint, it extracts the actual amount ratio of each key sub-item defined by the fingerprint in the announcement list, and compares it with the reasonable interval given by the fingerprint to obtain the deviation direction and intensity of the single item; then it is aggregated in the dimensions of professional category, construction stage or cost attribute to determine whether there is an abnormal concentration phenomenon, i.e. multiple key sub-items deviate significantly in the same direction, at the same level or in the same cost attribute, and outputs the deviation list, such as listing the sub-item identifier, hierarchical path, actual ratio, reference interval, deviation intensity and evidence source, to provide direct input for risk type and suggestion generation. The system comprehensively determines the structural deviation conclusion and the deviation intensity and concentration, and classifies and grades the risks according to the preset trigger rules: when the structural entropy deviation exceeds the threshold or there is an abnormal concentration of weight deviation, it outputs an uncomparable or induced low price risk prompt; the prompt content includes risk type, classification, trigger basis such as structural entropy deviation magnitude, number of deviated items and concentration dimension, list of key sub-items affected and hierarchical position, comparison summary with industry reference, and time stamp and version number, etc., to ensure that different links can be reviewed and traces can be left. For the high-risk nodes marked, the system generates executable processing suggestions based on industry reference and historical case library: for the over-splitting scene, it gives suggestions on merging or minimum measurement granularity, and clearly defines the measurement boundary and pricing range; for the abnormal merging scene, it suggests restoring the typical split granularity and listing the independent measurement rules; for the key sub-items with weight deviation, it provides suggestions on bid allocation, such as returning the measure fee / fixed cost from the ambiguous item to the corresponding process, and gives key points for clarification questions and answers, such as suggesting inquiring whether it is allowed to split the bid by xx granularity or whether the temporary items / odd items are listed separately, etc.); all suggestions are accompanied by evidence chains such as corresponding deviated items, reference interval sources and examples of similar projects to facilitate the bid team to quickly improve the bid strategy and clarification materials.The system falls the risk conclusion and the suggestion on the bid eligibility label, and outputs the bid eligibility prompt according to the enterprise's strategy threshold and resource preference; the bid eligibility pointer structure and weight are in a reasonable range or can be eliminated by lightweight clarification; the boundary bid eligibility pointer has a manageable structure or weight deviation, and the suggestion can be implemented in the pre-bid clarification; the structure and weight deviation of the non-bid eligibility pointer are serious, and have a substantial impact on the bid fairness or the estimated performance cost; the key impact items, the estimated impact on the bid and bid evaluation, and the recalculated conclusion after implementing the suggestion, such as whether the risk is downgraded after splitting / returning according to the suggestion, are displayed in the prompt to help the decision-making layer complete the selection and resource investment scheduling within a limited time.

[0077] It can be understood that in this way, excessive splitting, abnormal merging or weight abnormal concentration can be quantitatively identified and located before bidding, avoiding invalid investment caused by uncomparable prices found in the late bidding stage. Each risk conclusion is attached to a hierarchical path, deviation item and reference source to meet the explainability requirements of internal and external audits and external communication. Not only is the prompt risky, but also a bid allocation and clarification question and answer template consistent with industry practice is provided to directly serve pre-bid question answering and bid strategy optimization. By configuring threshold values, versioned reference libraries and recalculation paths, the conclusion can be quickly reviewed after clarification or version update. The two-dimensional review of structure and weight and the bid eligibility prompt are linked to the bidding recommendation and screening process to preferentially select bid documents with comparable prices and controllable risks, improve the bid hit rate, and reduce the probability of compliance disputes and performance deviations.

[0078] In some examples, the method further includes:

[0079] Extracting the adjacent and hierarchical sequence relationship of the sub-items based on the bill of quantities information;

[0080] Querying the adjacent transfer probability matrix fingerprint of the same type of list based on industry samples;

[0081] Constructing a transfer matrix according to the bill of quantities information and calculating the KL divergence with the adjacent transfer probability matrix fingerprint to judge the abnormal splitting of the abnormal splitting and prompt the bid eligibility of the current bid document.

[0082] It can be understood that the sub-items of the bill of quantities present a stable adjacent order and hierarchical connection mode: within the same specialty, sub-items often appear according to typical process logic, such as measurement and line laying to foundation trench excavation to earthwork outward transport / backfilling to compaction to foundation concrete, etc., which can be abstracted as a state-state transition relationship. The adjacent transition probability matrix obtained by statistical analysis of historical similar lists can be used as a reference for normal structures; when the target list has excessive splitting and insertion of a large number of scattered sub-items, or abnormal merging / rearrangement such as deletion or merging of typical nodes, dislocation of order, the observed transition matrix will deviate significantly from the fingerprint. By measuring the divergence such as KL divergence between the two and locating the transition edges with the largest contribution, it can be quantified and explained whether the structure is abnormal, thereby giving an objective suggestion on comparability and investability.

[0083] By way of example, after the layout restoration and field cleaning of the announcement list, the system can reconstruct the hierarchical tree of chapters-sections-subitems-subitems, and generate the adjacent sequence of the same layer under each parent node in line order, such as the horizontal adjacent relationship between brother items i and i+1, while recording the hierarchical connection sequence such as the closure of the parent and the first subitem, the closure of the subitem and the parent, the chapter transition across the parent node, etc. In order to reduce the pseudo-adjacent caused by layout noise, the system combines the subitem coding rules, the similarity of work content, and the consistency of the unit of measurement to check and correct the consistency of the adjacent relationship, such as page break or note line removal. For each adjacent edge, the edge weight information is saved, which is counted by default according to the frequency of occurrence, and can be optionally corrected by the amount ratio or the engineering quantity weight. The state space is unified to the standard term dictionary combined with the coding system of the hierarchical label, and the adjacent / hierarchical edge set that can be used for statistics is obtained. According to the project type, professional category and region, the system retrieves the industry benchmark library and historical winning samples, and extracts the same type of list set. For each list in the set, repeat the previous extraction process, and normalize the adjacent transition frequency in the same state space to obtain the adjacent transition probability matrix fingerprint. In order to improve robustness, the system performs smoothing processing on the rare transition with low support, and gives the confidence information of each typical transition, such as the number of projects, variance range, or typical upstream / downstream context. For term differences and granularity differences, the system unifies different expressions to a unified state through synonym mapping and hierarchical alignment, ensuring that the target list can be aligned one by one in the subsequent process. Finally, the same type of fingerprint matrix is formed, combined with the reference package of confidence interval and support, and cached for comparison and explanation. The system counts the adjacent / hierarchical edges obtained in the first step as the transition matrix of the target list, and completes the alignment in the same state space as the aforementioned fingerprint. For the unseen transition in the target matrix and the low-probability transition in the fingerprint, a consistent smoothing and truncation strategy is adopted to avoid extreme values. Then, the divergence measure of the two is calculated and the edge contribution is decomposed to locate the most inconsistent local structure, such as the diversion caused by inserting multiple rare subitems between earth excavation and external transport / backfill, or the missing of key transitions caused by abnormally combining templates, steel bars or concrete into one comprehensive item. According to the divergence value, the number of abnormal edges and the amount weight ratio, the system triggers abnormal splitting, abnormal merging or structure rearrangement risk, and marks the chapter and section where the anomaly is concentrated as a high-risk segment. Finally, the risk conclusion is mapped to the investability prompt for the current bid document, and structured evidence such as the list of abnormal edges, the hierarchical path, the associated amount ratio, or the comparison summary and timestamp with the fingerprint is output to support pre-bid communication and internal decision-making.

[0084] It can be understood that by indicating the process logic through adjacent or hierarchical transfer, whether it is intentionally split / combined or whether the sequence is abnormal is converted from subjective judgment to quantifiable divergence score, and the most contributing transfer edge and chapter are accurately positioned, facilitating quick clarification or adjustment of strategy. Using the fingerprints precipitated by the same list set, automatically adapting to different regional and industry terminology differences, significantly reducing misjudgment caused by different expressions. When the divergence anomaly is concentrated in high-amount chapters, the system can simultaneously prompt the risk level of non-comparable or induced low price, and return this level to the bidding information recommendation and screening strategy, to preferentially filter distorted projects, and improve hit rate and resource utilization rate.

[0085] According to some embodiments, further comprising:

[0086] Based on the bill of quantities information, extracting the work content description and measurement rules of each sub-item, and generating corresponding semantic vectors;

[0087] According to the semantic vectors, calculating the semantic similarity between sub-items to form a sub-item similarity matrix;

[0088] Based on the measurement rules of sub-items, calculating the range overlap coefficient between sub-items;

[0089] According to the sub-item similarity matrix and the range overlap coefficient, determining the overlap degree hot area and calculating the repeated measurement index;

[0090] In the case where the repeated measurement index exceeds the threshold, generating a repeated measurement prompt.

[0091] It can be understood that repeated measurement or range overlap in the bill of quantities is often hidden in text synonymous rewriting and measurement caliber boundaries: if only looking at line item names or simple keywords, it is easy to miss or misjudge. This method describes the closeness of sub-item work content in semantic space with semantic similarity, reflects the overlap degree of sub-item measurement caliber in space, process, time and material dimensions with range overlap coefficient, and uses both to identify overlap degree hot area and calculate repeated measurement index; When the index exceeds the threshold, output a structured repeated measurement prompt. It can avoid the shortcomings of single text matching, and avoid the distorted situation of only looking at quantity units while ignoring caliber boundaries, so as to more accurately locate whether there is a high-risk area of repeated measurement.

[0092] Exemplarily, the system can extract the work content description and measurement rule field line by line based on the bill of quantities, first do text normalization, such as professional term normalization, unit and symbol standardization, synonym and abbreviation expansion, and remove irrelevant modification, then encode the two types of text into semantic vectors in a double-channel way of content fragment combined with rule fragment; To improve robustness, the system performs clause segmentation on composite items and vectorizes them separately, while retaining hierarchical paths, item encoding, and measurement units, etc. metadata for subsequent alignment; For scanned copies or items with abnormal layout, the system repairs common broken phrases such as external transport or clean transport, backfill or backfill tamping after OCR through dictionary combined with context templates, and then vectorizes after ensuring the completeness of semantic load. The system calculates the semantic similarity between two items in the same parent node first, and then supplements the calculation across parent nodes but within the same professional chapter to obtain the global item similarity matrix; To reduce noise, the system sets weights for stop words, very short sentences, and templated descriptions, and introduces key phrase weighting based on industry word library such as transportation, loading and unloading, backfill, secondary handling, and temporary building demolition, etc. to make the similarity more focused on the core actions that can lead to repeated measurement; For large-scale lists, the system first uses the nearest neighbor index to do candidate recall, then calculates the similarity, and finally symmetrizes and truncates the matrix to avoid individual abnormal text dominating downstream judgment. Then the system extracts the measurement range boundary from the measurement rule and work content description, constructs a multi-dimensional range description, such as space or part, process or construction stage, material or equipment, time or progress window, measurement unit and conversion range, whether the clause is included or not; Then unify the units and convert the qualitative range description into comparable boundary labels and range dictionary, for example, soil disposal is transported to × × place, the transportation distance is within × × km or includes mechanical access to the site, and only once, etc. Assign values according to the three types of relationships of inclusion, intersection, and separation in the same dimension and get the range overlap coefficient by synthesis; When encountering unknown or contradictory clauses, the system labels them as uncertain boundaries and introduces conservative reduction in the coefficient to reduce the risk of false positives, while retaining this uncertainty as evidence for subsequent clarification questions. The system fuses the item similarity and range overlap coefficient obtained above pair by pair, and combines the item amount ratio or the size of the bill of quantities as business weight to get the comprehensive overlap strength of the item pair; Then threshold and cluster the overlap strength matrix to identify the overlap degree hotspots that appear in the same parent node or across adjacent chapters, for example, soil disposal-transportation of slag- disposal of soil are highly similar to each other and their ranges include each other; Aggregate the repeated measurement index at the hot zone level and output the core trigger factors of each hot zone, such as high similarity phrases, range clauses with the same or containing relationship, and amount ratio concentration, etc. to provide interpretable evidence for subsequent prompts and rectification suggestions.When the repeated measurement index exceeds the threshold value, the system generates a repeated measurement prompt, including: risk classification, associated sub-item list and hierarchical path, key evidence of similarity and overlap coefficient, involved measurement conflict and sample statement, estimated interval of affected amount or engineering quantity; At the same time, executable processing suggestions are given, such as merging similar items and clearly counting only once, splitting and clearly including transportation or handling, limiting the distance and disposal site range, or supplementing the secondary transportation pricing boundary; The prompt also gives the recalculation result after the suggested rectification, such as whether the hot area intensity and repeated measurement index are downgraded, and maps the conclusion to the current bid document as the bidability prompt and pre-bid clarification question points, supporting the bid team to make quick decisions and communicate.

[0093] It can be understood that, thereby, through the joint determination of semantic similarity and measurement range overlap, both implicit repetition caused by synonymous rewriting and false positives that look similar but do not overlap in range boundaries can be identified. Each risk is accompanied by a trigger phrase, range clause and hierarchical path, which can be directly converted into executable actions such as counting only once, range splitting or range merging, and the rectification effect can be verified by recalculation. The overlap intensity combined with the sub-item amount proportion or engineering quantity scale weighting makes the system preferentially prompt the repeated measurement that has a large impact on cost, improving the efficiency of pre-bid resources. The vocabulary, range dictionary and threshold can be dynamically updated according to the profession and region, adapting to different list styles and term habits to ensure stability and consistency in multiple industry scenarios. Moreover, repeated measurements are discovered and eliminated in the pre-bid stage, improving the uniformity of list range and the comparability of bid, reducing disputes and compliance risks caused by unclear range, thereby improving the overall bid certainty and project execution quality.

[0094] Please refer to Figure 2 An embodiment of the bidding information recommendation device in the embodiments of the present application can include:

[0095] The identification unit 21 is configured to identify key time node information of a target bid announcement, so as to construct a project qualification validity time axis in the case that the key time node information covers the registration deadline, the bid opening time, the contract signing and the planned completion.

[0096] The query unit 22 is configured to query the certificate validity period, the annual inspection and certificate replacement plan and the key personnel certificate state associated with the pre-stored qualification file matched by the bid subject.

[0097] The determination unit 23 is configured to determine the qualification coverage rate from the registration date to the planned completion date according to the time sequence matching of the project qualification validity time axis and the certificate validity period, and prompt the bidability of the current bid document.

[0098] In summary, the bidding information recommendation device provided by the embodiments of the present application identifies key time node information of a target bidding announcement, constructs a project qualification validity time axis in the case that the key time node information covers the registration deadline, the bid opening time, the contract signing and the planned completion, queries the certificate validity period, the annual inspection and the certificate replacement plan and the key personnel certificate status associated with the pre-stored qualification archives based on the pre-stored qualification archives matched by a bidding subject, performs time sequence matching according to the project qualification validity time axis and the certificate validity period, determines the qualification coverage rate from the registration day to the planned completion day, and prompts the bidability of the current bid document. Thus, for the inconsistency of the effectiveness window of the qualification certificate at different time points of registration, bid opening and performance, the conventional matching only checks whether the current one is effective, the present scheme can accurately identify the time sequence risks such as qualification invalidation on the bid opening day, annual inspection interruption during performance and personnel certificate replacement window, which are easily missed in traditional manual review. The qualification coverage rate measures the compliance degree throughout the process with a single numerical value, and the key node rigid checking is superimposed, which not only avoids the misjudgment of high coverage rate but invalidation at the key moment, but also avoids the optimistic deviation of node validity but mid-section interruption. The standardized time axis and interval segments make the prompt result traceable, accountable and reviewable, significantly reducing the communication cost and compliance disputes. In engineering practice, this determination strategy can also be linked with the recommendation system to preferentially push the bid document with high qualification continuity and safe key nodes, which empirically improves the bid hit rate and reduces the cost risk caused by invalid registration, flow bidding or performance obstruction.

[0099] As shown in Figure 3 The embodiments of the present application also provide an electronic device 300, which comprises a memory 310, a processor 320 and a computer program 311 stored in the memory 310 and capable of running on the processor, and the processor 320 implements the steps of any method of the above bidding information recommendation when executing the computer program 311.

[0100] Since the electronic device introduced in the embodiments is the device used to implement the bidding information recommendation device in the embodiments of the present application, the specific implementation mode of the electronic device and various changes thereof can be understood by those skilled in the art based on the method introduced in the embodiments of the present application, so that how the electronic device implements the method in the embodiments of the present application is not introduced in detail here, as long as the device used by those skilled in the art to implement the method in the embodiments of the present application belongs to the scope of the present application.

[0101] In the specific implementation process, the computer program 311 can implement Figure 1 any implementation mode in the corresponding embodiments when executed by the processor.

[0102] It should be noted that in the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0103] Those skilled in the art will appreciate that embodiments of the application can be provided as methods, systems, or computer program products. Accordingly, the application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the application can be embodied in the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk memory, CD-ROM, optical memory, etc.) having computer usable program code embodied therein.

[0104] The application is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded computer, or other programmable data processing apparatus to produce a machine, so that the instructions, which are executed via the processor of the computer or other programmable data processing apparatus, generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in the flowcharts and / or block diagrams.

[0105] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in the flowcharts and / or block diagrams.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in the flowcharts and / or block diagrams.

[0107] The embodiments of the application also provide a computer program product, which includes computer software instructions, when the computer software instructions run on a processing device, make the processing device execute the functions as described in the above embodiments. Figure 1The flow of the bidding information recommendation in the corresponding embodiment.

[0108] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the flow or function according to the embodiments of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that the computer can store or be integrated into a data storage device such as a server, data center, etc. containing one or more available media sets. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)) and the like.

[0109] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0110] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0111] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0112] In addition, each of the functional units in the various embodiments of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0113] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in part, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various other media that can store program codes.

[0114] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of recommending bidding information, characterized by, Comprise: Identify key time node information of the target bidding announcement to construct a project qualification validity timeline in the case of covering the registration deadline, bid opening time, contract signing and planned completion; Based on the pre-stored qualification file matched by the bidding subject, query the certificate validity period, annual inspection and replacement plan, and key personnel certificate status associated with the qualification file; According to the time sequence matching of the project qualification validity timeline and the certificate validity period, determine the qualification coverage rate from the registration date to the planned completion date, and prompt the bidability of the current bid.

2. The method of claim 1, wherein, Also include: Generate a certificate renewal buffer or personnel replacement suggestion if the coverage rate is below the preset threshold.

3. The method of claim 1, wherein, Also include: Obtain personnel and certificate pre-occupation time information of the bidding subject in parallel projects, and construct a cross-project qualification conflict graph in the case of covering the registration deadline, evaluation and performance key nodes of the current bid; Based on the pre-stored qualification file matched by the bidding subject, query the on-site commitment, shift plan and labor compliance restrictions associated with the qualification file; According to the time sequence matching of the conflict graph and the project qualification validity timeline, predict the cross-project qualification available time slot, generate a personnel scheduling or replacement scheme suggestion if the available time slot is insufficient, and prompt the bidability of the current bid.

4. The method of claim 1, wherein, Also include: Parse the bill of quantities information of the target bidding announcement; In the case of the bill of quantities information including itemized coding, unit of measurement and work content description, respectively extract the itemized level to generate an item-weight mapping table, the weight being the proportion of itemized amount to total amount, and the itemized amount being calculated based on quantity and unit price; Based on the matched industry benchmark list and historical winning samples, query the item fingerprint information corresponding to the announcement list, which contains the typical level depth, common item set and reference interval of each item weight of similar projects; According to the alignment result of the item fingerprint information and the announcement list, calculate the structural entropy and weight deviation rate using the item-weight mapping table to determine excessive splitting or abnormal merging, and prompt the bidability of the current bid.

5. The method of claim 4, wherein, The use of the item-weight mapping table to calculate the structural entropy and weight deviation rate to determine excessive splitting or abnormal merging, and prompt the bidability of the current bid, comprises: In the case of structural entropy deviating from the preset threshold or weight deviation rate being abnormally concentrated, generate an incomparable price or induced low price risk prompt; Output the bid allocation and clarification question and answer suggestions of the key items.

6. The method of claim 4, wherein, Also include: Extract the adjacent and hierarchical sequence relationship of the items based on the bill of quantities information; Query the adjacent transfer probability matrix fingerprint of similar lists based on industry samples; According to the bill of quantities information, construct a transfer matrix and calculate the KL divergence with the adjacent transfer probability matrix fingerprint to determine abnormal splitting in the case of divergence anomaly, and prompt the bidability of the current bid.

7. The method of claim 4, wherein, Also include: Based on the bill of quantities information, extract the work content description and measurement rules of each item to generate corresponding semantic vectors; According to the semantic vectors, calculate the semantic similarity between items to form an item similarity matrix; A range overlap coefficient between the items is calculated based on the item-based metrology rules; An overlap degree hot area is determined according to the item similarity matrix and the range overlap coefficient, and a duplicate metrology index is calculated; In the case where the duplicate metrology index exceeds a threshold value, a duplicate metrology prompt is generated.

8. A bid information recommendation device characterized by comprising: Comprise: An identifying unit is configured to identify key time node information of a target bidding announcement, so as to construct a project qualification validity time axis in the case where the key time node information covers a bid submission deadline, a bid opening time, a contract signing and a planned completion; A querying unit is configured to query a certificate validity period, an annual inspection and a certificate replacement plan and a key personnel certificate state associated with a pre-stored qualification file based on the pre-stored qualification file matched by a bidding subject; A determining unit is configured to determine a qualification coverage rate from a bid submission day to a planned completion day according to a time sequence matching between the project qualification validity time axis and the certificate validity period, and to prompt a bidability of a current bid.

9. An electronic device comprising: A memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the bidding information recommendation method according to any one of claims 1-7 when executing the computer program stored in the memory.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the bidding information recommendation method according to any one of claims 1-7.

Images