A semiconductor data testing method and test abnormality warning system

Abstract

The present invention discloses a semiconductor data test method and a test abnormality warning system, which belongs to the field of semiconductor data test technology, and includes the following steps: S1: Based on the semiconductor data test abnormality warning demand, determine the semiconductor data test information; S2: Based on the sequential retrieval method and feature extraction, determine the semiconductor test feature data; S3: Based on the semiconductor test standard data, compare and analyze the semiconductor test feature data to determine the corresponding semiconductor data test analysis results; S4: Based on the semiconductor data test abnormality warning method, perform abnormal warning on the semiconductor data test. The present invention solves the existing problem that the semiconductor data test cannot be abnormally warned, resulting in poor semiconductor data test effect. The present invention can perform abnormal warning on the semiconductor data test, so that the tester can understand the semiconductor data test situation in time and improve the semiconductor data test effect.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor data testing technology, and in particular to a semiconductor data testing method and a test abnormality early warning system. Background Art

[0002] The semiconductor production process is as follows: it consists of wafer manufacturing, wafer testing, chip packaging and post-packaging testing; semiconductor packaging testing refers to the process of processing the tested wafers into independent chips according to product models and functional requirements.

[0003] The Chinese patent with publication number CN116303696B discloses a method, device, electronic device and storage medium for processing semiconductor test data. Through aggregation processing, elimination of abnormal test data and stratified sampling, the load pressure of the application in the semiconductor data scatter plot analysis scenario is reduced in a hierarchical and multi-step manner while maintaining no significant difference between the sample and the original data, meeting the scenario's focus on overall data distribution and boundary values, and improving the reliability of the overall system; however, the patent has the following defects:

[0004] After the existing semiconductor data test, no abnormal warning can be given for the semiconductor data test, resulting in the tester being unable to understand the semiconductor data test situation in a timely manner, resulting in poor semiconductor data test results. Summary of the invention

[0005] The purpose of the present invention is to provide a semiconductor data testing method and a test abnormality warning system, which can provide abnormal warning for semiconductor data testing, so that testers can understand the semiconductor data testing situation in time, improve the semiconductor data testing effect, and solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A semiconductor data testing method comprises the following steps:

[0008] S1: Based on the semiconductor data test abnormality warning requirement, a semiconductor data test information request instruction is sent to the test terminal, the test terminal transmits the semiconductor data test information to the mobile terminal, the mobile terminal obtains the semiconductor data test information, and determines the semiconductor data test information;

[0009] S2: Based on the sequential retrieval method, the semiconductor data test information is retrieved to determine the semiconductor data test information that is valuable for the semiconductor data test abnormality warning, and the semiconductor data test information is feature extracted to determine the semiconductor test feature data;

[0010] S3: Based on the semiconductor test characteristic data, index the semiconductor test standard data that matches the semiconductor test characteristic data, retrieve the semiconductor test standard data, and compare and analyze the semiconductor test characteristic data based on the semiconductor test standard data to determine the corresponding semiconductor data test analysis result;

[0011] S4: Based on data mining technology, conduct in-depth mining and analysis on the semiconductor data test analysis results, determine the corresponding semiconductor data test abnormality warning method, and perform abnormal warning on the semiconductor data test based on the semiconductor data test abnormality warning method.

[0012] Preferably, in S1, determining the semiconductor data test information includes:

[0013] Based on the abnormal warning requirements of semiconductor data testing;

[0014] The mobile terminal sends a request for semiconductor data test information instruction to the test terminal;

[0015] The test terminal receives the instruction for requesting semiconductor data test information sent by the mobile terminal, the test terminal responds in a timely manner, and the test terminal transmits the semiconductor data test information to the mobile terminal;

[0016] The mobile terminal receives the semiconductor data test information transmitted by the test terminal, obtains the semiconductor data test information, and determines the semiconductor data test information.

[0017] Preferably, in S3, based on the semiconductor test standard data, the semiconductor test characteristic data is compared and analyzed to determine the corresponding semiconductor data test analysis result, including:

[0018] Acquire semiconductor test characteristic data;

[0019] Based on the semiconductor test characteristic data, indexing semiconductor test standard data matching the semiconductor test characteristic data, and determining the indexed semiconductor test standard data;

[0020] Semiconductor test standard data based on the index, and the semiconductor test standard data is retrieved;

[0021] Based on semiconductor test standard data, compare and analyze semiconductor test characteristic data to determine the corresponding semiconductor data test analysis results;

[0022] In the case where the semiconductor test characteristic data matches the semiconductor test standard data, the corresponding semiconductor data test analysis result is determined to be normal for the semiconductor data test, and there is no need to issue an abnormal warning for the semiconductor data test;

[0023] In the case where the semiconductor test characteristic data does not match the semiconductor test standard data, the corresponding semiconductor data test analysis result is determined to be a semiconductor data test anomaly, and an abnormal warning for the semiconductor data test is required.

[0024] Preferably, in S4, performing abnormal warning on semiconductor data test based on the semiconductor data test abnormal warning method includes:

[0025] Obtain semiconductor data test analysis results;

[0026] Based on data mining technology, the semiconductor data test analysis results are deeply mined and analyzed to determine the corresponding semiconductor data test abnormality warning method;

[0027] In the case where the semiconductor data test analysis result shows that the semiconductor data test is normal and there is no need to conduct abnormal warning for the semiconductor data test, the corresponding semiconductor data test abnormal warning method is to conduct normal testing on subsequent semiconductor data and continuously monitor the semiconductor data test results;

[0028] In the case where the semiconductor data test analysis result is that the semiconductor data test is abnormal and an abnormal warning is required for the semiconductor data test, the corresponding semiconductor data test abnormal warning method is determined to be an abnormal warning for the current semiconductor data test result, and based on the in-depth mining and analysis of the semiconductor data test analysis results, the corresponding cause of the semiconductor data test abnormality is determined, and the cause of the semiconductor data test abnormality is transmitted to the semiconductor tester for warning the semiconductor tester to manage the semiconductor test.

[0029] Preferably, according to the in-depth mining analysis of the semiconductor data test analysis results, the corresponding semiconductor data test abnormality causes are determined, including:

[0030] Acquire abnormal semiconductor data that does not match semiconductor test standard data based on semiconductor data test analysis results, and determine data difference characteristics between the abnormal semiconductor data and the semiconductor test standard data;

[0031] Inputting the semiconductor data type and semiconductor model data corresponding to the abnormal semiconductor data into a pre-designed abnormality identification and determination model to preliminarily obtain the possible first abnormal cause;

[0032] Acquire an associated data difference feature associated with the data difference feature based on the abnormal feature map, acquire a first abnormal feature corresponding to the data difference feature, and acquire a second abnormal feature corresponding to the associated data difference feature;

[0033] Acquire a standard abnormal feature corresponding to the first abnormal cause based on the abnormal feature map;

[0034] Setting a first weight for the first abnormal feature and a second weight for the second abnormal feature, and obtaining a first weighted abnormal feature and a second weighted abnormal feature based on the first weight and the second weight;

[0035] Perform weighted matching on the first weighted abnormal feature and the second weighted abnormal feature with the standard abnormal feature to obtain a first target standard abnormal feature having a matching degree greater than a preset matching degree, and obtain a second abnormal cause corresponding to the first target standard abnormal feature;

[0036] The standard abnormal feature corresponding to the second abnormal cause is matched with the first abnormal feature to obtain the second target standard abnormal feature with the highest matching degree, and the target abnormal cause corresponding to the second target standard abnormal feature is obtained, and the target abnormal cause is the corresponding semiconductor data test abnormal cause.

[0037] According to another aspect of the present invention, a semiconductor data test abnormality warning system is provided, which is used to implement a semiconductor data test method as described above, comprising:

[0038] An information acquisition module is used to acquire semiconductor test records of semiconductor data tests performed on a semiconductor test bench to determine semiconductor data test information;

[0039] An information processing module, used for processing semiconductor data test information to determine semiconductor test characteristic data;

[0040] A data analysis module is used to analyze semiconductor test characteristic data and determine semiconductor data test analysis results;

[0041] The abnormal warning module is used to issue abnormal warnings for semiconductor data test conditions and determine corresponding semiconductor data test abnormal warning methods.

[0042] Preferably, the information acquisition module includes:

[0043] An information request unit, used for requesting semiconductor data test information from the test terminal;

[0044] Based on the semiconductor data test abnormal warning demand, send a semiconductor data test information request instruction to the test terminal, requesting the test terminal to transmit the semiconductor data test information;

[0045] An information transmission unit, used for transmitting semiconductor data test information;

[0046] After the test terminal receives the semiconductor data test information request instruction sent by the mobile terminal, the test terminal transmits the semiconductor data test information to the mobile terminal;

[0047] An information acquisition unit, used for acquiring semiconductor data test information;

[0048] The mobile terminal receives the semiconductor data test information transmitted by the test terminal, obtains the semiconductor data test information, and determines the semiconductor data test information.

[0049] The information processing module includes:

[0050] An information retrieval unit, used for retrieving semiconductor data test information;

[0051] Acquire semiconductor data test information, and search the semiconductor data test information based on a sequential search method;

[0052] Filter out semiconductor data test information that is not valuable for semiconductor data test abnormality warning, and determine semiconductor data test information that is valuable for semiconductor data test abnormality warning;

[0053] A feature extraction unit, used for extracting features from semiconductor data test information;

[0054] Semiconductor data test information valuable for semiconductor data test abnormality warning is obtained, and features are extracted from the semiconductor data test information to determine semiconductor test feature data.

[0055] Preferably, the data analysis module includes:

[0056] A data storage unit, used to store pre-set semiconductor test standard data, providing a reference and guidance basis for semiconductor data testing;

[0057] A data indexing unit, used for indexing pre-set semiconductor test standard data;

[0058] Acquire semiconductor test characteristic data, and based on the semiconductor test characteristic data, index semiconductor test standard data matching the semiconductor test characteristic data from a data storage unit, and determine the indexed semiconductor test standard data;

[0059] A data retrieval unit, used to retrieve the indexed semiconductor test standard data;

[0060] Obtaining the indexed semiconductor test standard data, based on the semiconductor test standard data, and retrieving the semiconductor test standard data;

[0061] A comparison and analysis unit, used for comparing and analyzing semiconductor test characteristic data;

[0062] Acquire semiconductor test characteristic data and semiconductor test standard data, perform comparative analysis on the semiconductor test characteristic data based on the semiconductor test standard data, and determine corresponding semiconductor data test analysis results.

[0063] Preferably, the comparison and analysis unit comprises:

[0064] A standardization unit, used for performing data standardization on semiconductor test characteristic data according to data format characteristics of semiconductor test standard data to obtain target semiconductor test characteristic data;

[0065] A data analysis unit, used to calculate the data importance P of the semiconductor test standard data based on the participation weight of the data type in the standard test result and according to the following formula;

[0066]

[0067] Where n represents the number of data types of semiconductor test standard data, N i represents the data volume of the i-th data type in the semiconductor test standard data, ε i represents the participation weight of the i-th data type in the semiconductor test standard data, θ represents the test difficulty of the standard test result, D j represents the amount of data of the jth data type in the standard test results, ε j Indicates the participation weight of the jth data type in the standard test results;

[0068] A comparison unit, used for calculating the comparison similarity S between the target semiconductor test feature data and the semiconductor test standard data based on the data importance of the semiconductor test standard data and according to the following formula;

[0069] S=P*[1-|K 0 -K a |]

[0070] Among them, K 0 Represents the data characteristics of semiconductor test standard data, K a data features representing target semiconductor test feature data;

[0071] The result determination unit is used to determine the corresponding semiconductor data test analysis result based on the comparison similarity between the target semiconductor test characteristic data and the semiconductor test standard data.

[0072] Preferably, the abnormal warning module includes:

[0073] A mining and analysis unit, used for mining and analyzing semiconductor data test analysis results;

[0074] Obtain semiconductor data test analysis results, conduct in-depth mining and analysis on the semiconductor data test analysis results based on data mining technology, and determine the corresponding semiconductor data test abnormality warning method;

[0075] An abnormal warning unit is used to issue an abnormal warning for semiconductor data test conditions;

[0076] A semiconductor data test abnormality warning method is obtained, and an abnormality warning is performed on the semiconductor data test based on the semiconductor data test abnormality warning method.

[0077] Compared with the prior art, the present invention has the following beneficial effects:

[0078] The semiconductor data testing method and test abnormality warning system of the present invention determine semiconductor data testing information based on semiconductor data testing abnormality warning requirements, and process the semiconductor data testing information to determine semiconductor test characteristic data. Based on semiconductor test standard data, the semiconductor test characteristic data is compared and analyzed to determine corresponding semiconductor data test analysis results. Based on data mining technology, the semiconductor data test analysis results are deeply mined and analyzed to determine the corresponding semiconductor data test abnormality warning method. Based on the semiconductor data test abnormality warning method, abnormal warning is performed on semiconductor data testing. Abnormal warning can be performed on semiconductor data testing, so that testers can understand the semiconductor data test situation in time and improve the semiconductor data test effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 A flowchart of a semiconductor data testing method of the present invention;

[0080] Figure 2 This is a module schematic diagram of the semiconductor data test abnormality warning system of the present invention. DETAILED DESCRIPTION

[0081] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0082] In order to solve the problem that after the existing semiconductor data test, the semiconductor data test cannot be abnormally warned, resulting in the tester not being able to understand the semiconductor data test situation in time, resulting in poor semiconductor data test results, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0083] A semiconductor data testing method comprises the following steps:

[0084] S1: Based on the semiconductor data test abnormality warning requirement, a semiconductor data test information request instruction is sent to the test terminal, the test terminal transmits the semiconductor data test information to the mobile terminal, the mobile terminal obtains the semiconductor data test information, and determines the semiconductor data test information;

[0085] As a preferred technical solution of the present invention, the determination of semiconductor data test information includes:

[0086] Based on the abnormal warning requirements of semiconductor data testing;

[0087] The mobile terminal sends a request for semiconductor data test information instruction to the test terminal;

[0088] The test terminal receives the instruction for requesting semiconductor data test information sent by the mobile terminal, the test terminal responds in a timely manner, and the test terminal transmits the semiconductor data test information to the mobile terminal;

[0089] The mobile terminal receives the semiconductor data test information transmitted by the test terminal, obtains the semiconductor data test information, and determines the semiconductor data test information.

[0090] S2: Based on the sequential retrieval method, the semiconductor data test information is retrieved to determine the semiconductor data test information that is valuable for the semiconductor data test abnormality warning, and the semiconductor data test information is feature extracted to determine the semiconductor test feature data;

[0091] S3: Based on the semiconductor test characteristic data, index the semiconductor test standard data that matches the semiconductor test characteristic data, retrieve the semiconductor test standard data, and compare and analyze the semiconductor test characteristic data based on the semiconductor test standard data to determine the corresponding semiconductor data test analysis result;

[0092] As a preferred technical solution of the present invention, the determination of the semiconductor data test analysis result includes:

[0093] Acquire semiconductor test characteristic data;

[0094] Based on the semiconductor test characteristic data, indexing semiconductor test standard data matching the semiconductor test characteristic data, and determining the indexed semiconductor test standard data;

[0095] Semiconductor test standard data based on the index, and the semiconductor test standard data is retrieved;

[0096] Based on semiconductor test standard data, compare and analyze semiconductor test characteristic data to determine the corresponding semiconductor data test analysis results;

[0097] In the case where the semiconductor test characteristic data matches the semiconductor test standard data, the corresponding semiconductor data test analysis result is determined to be normal for the semiconductor data test, and there is no need to issue an abnormal warning for the semiconductor data test;

[0098] In the case where the semiconductor test characteristic data does not match the semiconductor test standard data, the corresponding semiconductor data test analysis result is determined to be a semiconductor data test anomaly, and an abnormal warning for the semiconductor data test is required.

[0099] S4: Based on data mining technology, conduct in-depth mining and analysis on the semiconductor data test analysis results, determine the corresponding semiconductor data test abnormality warning method, and perform abnormal warning on the semiconductor data test based on the semiconductor data test abnormality warning method.

[0100] As a preferred technical solution of the present invention, an abnormal warning is performed on semiconductor data test based on a semiconductor data test abnormal warning method, including:

[0101] Obtain semiconductor data test analysis results;

[0102] Based on data mining technology, the semiconductor data test analysis results are deeply mined and analyzed to determine the corresponding semiconductor data test abnormality warning method;

[0103] In the case where the semiconductor data test analysis result shows that the semiconductor data test is normal and there is no need to conduct abnormal warning for the semiconductor data test, the corresponding semiconductor data test abnormal warning method is to conduct normal testing on subsequent semiconductor data and continuously monitor the semiconductor data test results;

[0104] In the case where the semiconductor data test analysis result is that the semiconductor data test is abnormal and an abnormal warning is required for the semiconductor data test, the corresponding semiconductor data test abnormal warning method is determined to be an abnormal warning for the current semiconductor data test result, and based on the in-depth mining and analysis of the semiconductor data test analysis results, the corresponding cause of the semiconductor data test abnormality is determined, and the cause of the semiconductor data test abnormality is transmitted to the semiconductor tester for warning the semiconductor tester to manage the semiconductor test.

[0105] In one embodiment, according to the deep mining analysis of the semiconductor data test analysis results, the corresponding semiconductor data test abnormality causes are determined, including:

[0106] Acquire abnormal semiconductor data that does not match semiconductor test standard data based on semiconductor data test analysis results, and determine data difference characteristics between the abnormal semiconductor data and the semiconductor test standard data;

[0107] Inputting the semiconductor data type and semiconductor model data corresponding to the abnormal semiconductor data into a pre-designed abnormality identification and determination model to preliminarily obtain the possible first abnormal cause;

[0108] Acquire an associated data difference feature associated with the data difference feature based on the abnormal feature map, acquire a first abnormal feature corresponding to the data difference feature, and acquire a second abnormal feature corresponding to the associated data difference feature;

[0109] Acquire a standard abnormal feature corresponding to the first abnormal cause based on the abnormal feature map;

[0110] Setting a first weight for the first abnormal feature and a second weight for the second abnormal feature, and obtaining a first weighted abnormal feature and a second weighted abnormal feature based on the first weight and the second weight;

[0111] Perform weighted matching on the first weighted abnormal feature and the second weighted abnormal feature with the standard abnormal feature to obtain a first target standard abnormal feature having a matching degree greater than a preset matching degree, and obtain a second abnormal cause corresponding to the first target standard abnormal feature;

[0112] The standard abnormal feature corresponding to the second abnormal cause is matched with the first abnormal feature to obtain the second target standard abnormal feature with the highest matching degree, and the target abnormal cause corresponding to the second target standard abnormal feature is obtained, and the target abnormal cause is the corresponding semiconductor data test abnormal cause.

[0113] In this embodiment, there are multiple first abnormal causes.

[0114] In this embodiment, the first weight is greater than the second weight.

[0115] In this embodiment, there are multiple second abnormal causes.

[0116] The beneficial effects of the above design scheme are: by obtaining abnormal semiconductor data that does not match the semiconductor test standard data based on the semiconductor data test analysis results, and determining the data difference characteristics between the abnormal semiconductor data and the semiconductor test standard data; inputting the semiconductor data type and semiconductor model data corresponding to the abnormal semiconductor data into a pre-designed abnormal identification and determination model, preliminarily obtaining the possible first abnormal cause, realizing the preliminary identification and investigation of the abnormal cause, and providing a basis for subsequent accurate investigation; secondly, setting a first weight for the first abnormal feature and a second weight for the second abnormal feature, and obtaining a first weighted abnormal feature and a second weighted abnormal feature based on the first weight and the second weight; comparing the first weighted abnormal feature and the second weighted abnormal feature with the standard data; The quasi-abnormal features are weighted matched to obtain the first target standard abnormal feature with a matching degree greater than the preset matching degree, and the second abnormal cause corresponding to the first target standard abnormal feature is obtained. The abnormal cause is weighted matched by combining the abnormal feature and its associated associated abnormal feature to obtain the second abnormal cause, and the accuracy of the obtained second abnormal cause is guaranteed. Finally, the standard abnormal feature corresponding to the second abnormal cause is matched with the first abnormal feature to obtain the second target standard abnormal feature with the highest matching degree, and the target abnormal cause corresponding to the second target standard abnormal feature is obtained. The target abnormal cause is the corresponding semiconductor data test abnormal cause, which ensures the optimality and accuracy of the final semiconductor data test abnormal cause and improves the semiconductor data test effect.

[0117] In order to better demonstrate the semiconductor data test abnormality warning process, this embodiment now proposes a semiconductor data test abnormality warning system, which is used to implement a semiconductor data test method as described above, including:

[0118] An information acquisition module is used to acquire semiconductor test records of semiconductor data tests performed on a semiconductor test bench to determine semiconductor data test information;

[0119] As a preferred technical solution of the present invention, the information acquisition module includes:

[0120] An information request unit, used for requesting semiconductor data test information from the test terminal;

[0121] Based on the semiconductor data test abnormal warning demand, send a semiconductor data test information request instruction to the test terminal, requesting the test terminal to transmit the semiconductor data test information;

[0122] An information transmission unit, used for transmitting semiconductor data test information;

[0123] After the test terminal receives the semiconductor data test information request instruction sent by the mobile terminal, the test terminal transmits the semiconductor data test information to the mobile terminal;

[0124] An information acquisition unit, used for acquiring semiconductor data test information;

[0125] The mobile terminal receives the semiconductor data test information transmitted by the test terminal, obtains the semiconductor data test information, and determines the semiconductor data test information.

[0126] An information processing module, used for processing semiconductor data test information to determine semiconductor test characteristic data;

[0127] As a preferred technical solution of the present invention, the information processing module includes:

[0128] An information retrieval unit, used for retrieving semiconductor data test information;

[0129] Acquire semiconductor data test information, and search the semiconductor data test information based on a sequential search method;

[0130] Filter out semiconductor data test information that is not valuable for semiconductor data test abnormality warning, and determine semiconductor data test information that is valuable for semiconductor data test abnormality warning;

[0131] A feature extraction unit, used for extracting features from semiconductor data test information;

[0132] Semiconductor data test information valuable for semiconductor data test abnormality warning is obtained, and features are extracted from the semiconductor data test information to determine semiconductor test feature data.

[0133] A data analysis module is used to analyze semiconductor test characteristic data and determine semiconductor data test analysis results;

[0134] As a preferred technical solution of the present invention, the data analysis module includes:

[0135] A data storage unit, used to store pre-set semiconductor test standard data, providing a reference and guidance basis for semiconductor data testing;

[0136] A data indexing unit, used for indexing pre-set semiconductor test standard data;

[0137] Acquire semiconductor test characteristic data, and based on the semiconductor test characteristic data, index semiconductor test standard data matching the semiconductor test characteristic data from a data storage unit, and determine the indexed semiconductor test standard data;

[0138] A data retrieval unit, used to retrieve the indexed semiconductor test standard data;

[0139] Obtaining the indexed semiconductor test standard data, based on the semiconductor test standard data, and retrieving the semiconductor test standard data;

[0140] A comparison and analysis unit, used for comparing and analyzing semiconductor test characteristic data;

[0141] Acquire semiconductor test characteristic data and semiconductor test standard data, perform comparative analysis on the semiconductor test characteristic data based on the semiconductor test standard data, and determine corresponding semiconductor data test analysis results.

[0142] In one embodiment, the comparison and analysis unit includes:

[0143] A standardization unit, used for performing data standardization on semiconductor test characteristic data according to data format characteristics of semiconductor test standard data to obtain target semiconductor test characteristic data;

[0144] A data analysis unit, used to calculate the data importance P of the semiconductor test standard data based on the participation weight of the data type in the standard test result and according to the following formula;

[0145]

[0146] Where n represents the number of data types of semiconductor test standard data, N i represents the data volume of the i-th data type in the semiconductor test standard data, ε i represents the participation weight of the i-th data type in the semiconductor test standard data, θ represents the test difficulty of the standard test result, D j represents the amount of data of the jth data type in the standard test results, ε j represents the participation weight of the jth data type in the standard test result, and m represents the number of all data types in the standard test result;

[0147] A comparison unit, used for calculating the comparison similarity S between the target semiconductor test feature data and the semiconductor test standard data based on the data importance of the semiconductor test standard data and according to the following formula;

[0148] S=P*[1-|K 0 -K a |]

[0149] Among them, K 0 Represents the data characteristics of semiconductor test standard data, K a data features representing target semiconductor test feature data;

[0150] The result determination unit is used to determine the corresponding semiconductor data test analysis result based on the comparison similarity between the target semiconductor test characteristic data and the semiconductor test standard data.

[0151] In this embodiment, the standard test results are pre-designed based on historical test results.

[0152] In this embodiment, the greater the influence of the data type on the determination of the standard test result, the greater the participation weight of the data type in the standard test result.

[0153] In this embodiment, the values ​​of the data features of the semiconductor test standard data and the data features of the target semiconductor test feature data are predefined, and the value range is (0, 1), |K 0 -K a The smaller | is, the higher the similarity is.

[0154] The beneficial effects of the above design scheme are: target semiconductor test feature data is obtained by standardizing the semiconductor test feature data according to the data format characteristics of the semiconductor test standard data, providing a consistent data format basis for comparison between data, calculating the data importance of the semiconductor test standard data based on the participation weights of the data types in the standard test results, providing a weighted basis for further similarity comparison between data, highlighting the importance of important data in the analysis results, calculating the comparative similarity between the target semiconductor test feature data and the semiconductor test standard data based on the data importance of the semiconductor test standard data, and determining the corresponding semiconductor data test analysis results based on the comparative similarity between the target semiconductor test feature data and the semiconductor test standard data, thereby ensuring the accuracy of the obtained semiconductor data test analysis results.

[0155] The abnormal warning module is used to issue abnormal warnings for semiconductor data test conditions and determine corresponding semiconductor data test abnormal warning methods.

[0156] As a preferred technical solution of the present invention, the abnormal warning module includes:

[0157] A mining and analysis unit, used for mining and analyzing semiconductor data test analysis results;

[0158] Obtain semiconductor data test analysis results, conduct in-depth mining and analysis on the semiconductor data test analysis results based on data mining technology, and determine the corresponding semiconductor data test abnormality warning method;

[0159] An abnormal warning unit is used to issue an abnormal warning for semiconductor data test conditions;

[0160] A semiconductor data test abnormality warning method is obtained, and an abnormality warning is performed on the semiconductor data test based on the semiconductor data test abnormality warning method.

[0161] In summary, the semiconductor data testing method and test abnormality warning system of the present invention determine the semiconductor data test information based on the semiconductor data test abnormality warning demand, and process the semiconductor data test information to determine the semiconductor test characteristic data, and compare and analyze the semiconductor test characteristic data based on the semiconductor test standard data to determine the corresponding semiconductor data test analysis results, and conduct in-depth mining and analysis on the semiconductor data test analysis results based on data mining technology to determine the corresponding semiconductor data test abnormality warning method, and perform abnormality warning on the semiconductor data test based on the semiconductor data test abnormality warning method. Abnormal warning can be performed on the semiconductor data test, so that the test personnel can understand the semiconductor data test situation in time and improve the semiconductor data test effect.

[0162] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0163] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A semiconductor data testing method, characterized in that: The steps include: S1: Based on the semiconductor data test abnormality warning requirement, a semiconductor data test information request instruction is sent to the test terminal, the test terminal transmits the semiconductor data test information to the mobile terminal, the mobile terminal obtains the semiconductor data test information, and determines the semiconductor data test information; S2: Based on the sequential retrieval method, the semiconductor data test information is retrieved to determine the semiconductor data test information that is valuable for the semiconductor data test abnormality warning, and the semiconductor data test information is feature extracted to determine the semiconductor test feature data; S3: Based on the semiconductor test characteristic data, index the semiconductor test standard data that matches the semiconductor test characteristic data, retrieve the semiconductor test standard data, and compare and analyze the semiconductor test characteristic data based on the semiconductor test standard data to determine the corresponding semiconductor data test analysis result; S4: Based on data mining technology, perform in-depth mining and analysis on the semiconductor data test analysis results, determine the corresponding semiconductor data test abnormality warning method, and perform abnormal warning on the semiconductor data test based on the semiconductor data test abnormality warning method; In S4, an abnormal warning is performed on the semiconductor data test based on the semiconductor data test abnormal warning method, including: Obtain semiconductor data test analysis results; Based on data mining technology, the semiconductor data test analysis results are deeply mined and analyzed to determine the corresponding semiconductor data test abnormality warning method; In the case where the semiconductor data test analysis result shows that the semiconductor data test is normal and there is no need to perform abnormal warning on the semiconductor data test, the corresponding semiconductor data test abnormal warning method is to test the subsequent semiconductor data normally and continuously monitor the semiconductor data test results; In case that the semiconductor data test analysis result is that the semiconductor data test is abnormal and an abnormal warning for the semiconductor data test is required, the corresponding semiconductor data test abnormal warning method is determined to be an abnormal warning for the current semiconductor data test result, and based on the in-depth mining analysis of the semiconductor data test analysis result, the corresponding semiconductor data test abnormal cause is determined, and the semiconductor data test abnormal cause is transmitted to the semiconductor tester for warning the semiconductor tester to manage the semiconductor test; Based on the in-depth mining and analysis of the semiconductor data test analysis results, the corresponding semiconductor data test abnormal causes are determined, including: Acquire abnormal semiconductor data that does not match semiconductor test standard data based on semiconductor data test analysis results, and determine data difference characteristics between the abnormal semiconductor data and the semiconductor test standard data; Inputting the semiconductor data type and semiconductor model data corresponding to the abnormal semiconductor data into a pre-designed abnormality identification and determination model to preliminarily obtain the possible first abnormal cause; Acquire an associated data difference feature associated with the data difference feature based on the abnormal feature map, acquire a first abnormal feature corresponding to the data difference feature, and acquire a second abnormal feature corresponding to the associated data difference feature; Acquire a standard abnormal feature corresponding to the first abnormal cause based on the abnormal feature map; Setting a first weight for the first abnormal feature and a second weight for the second abnormal feature, and obtaining a first weighted abnormal feature and a second weighted abnormal feature based on the first weight and the second weight; Perform weighted matching on the first weighted abnormal feature and the second weighted abnormal feature with the standard abnormal feature to obtain a first target standard abnormal feature having a matching degree greater than a preset matching degree, and obtain a second abnormal cause corresponding to the first target standard abnormal feature; The standard abnormal feature corresponding to the second abnormal cause is matched with the first abnormal feature to obtain the second target standard abnormal feature with the highest matching degree, and the target abnormal cause corresponding to the second target standard abnormal feature is obtained, and the target abnormal cause is the corresponding semiconductor data test abnormal cause.

2. A semiconductor data testing method according to claim 1, characterized in that: In S1, semiconductor data test information is determined, including: Based on the abnormal warning requirements of semiconductor data testing; The mobile terminal sends a request for semiconductor data test information instruction to the test terminal; The test terminal receives the instruction for requesting semiconductor data test information sent by the mobile terminal, the test terminal responds in a timely manner, and the test terminal transmits the semiconductor data test information to the mobile terminal; The mobile terminal receives the semiconductor data test information transmitted by the test terminal, obtains the semiconductor data test information, and determines the semiconductor data test information.

3. A semiconductor data testing method according to claim 1, characterized in that: In S3, based on the semiconductor test standard data, the semiconductor test characteristic data is compared and analyzed to determine the corresponding semiconductor data test analysis results, including: Acquire semiconductor test characteristic data; Based on the semiconductor test characteristic data, indexing semiconductor test standard data matching the semiconductor test characteristic data, and determining the indexed semiconductor test standard data; Semiconductor test standard data based on the index, and the semiconductor test standard data is retrieved; Based on semiconductor test standard data, compare and analyze semiconductor test characteristic data to determine the corresponding semiconductor data test analysis results; In the case where the semiconductor test characteristic data matches the semiconductor test standard data, the corresponding semiconductor data test analysis result is determined to be normal for the semiconductor data test, and there is no need to issue an abnormal warning for the semiconductor data test; In the case where the semiconductor test characteristic data does not match the semiconductor test standard data, the corresponding semiconductor data test analysis result is determined to be a semiconductor data test anomaly, and an abnormal warning for the semiconductor data test is required.

4. A semiconductor data test abnormality warning system, used to implement the semiconductor data test method according to any one of claims 1 to 3, characterized in that: include: An information acquisition module is used to acquire semiconductor test records of semiconductor data tests performed on a semiconductor test bench to determine semiconductor data test information; An information processing module, used for processing semiconductor data test information to determine semiconductor test characteristic data; A data analysis module is used to analyze semiconductor test characteristic data and determine semiconductor data test analysis results; The abnormal warning module is used to issue abnormal warnings for semiconductor data test conditions and determine corresponding semiconductor data test abnormal warning methods.

5. A semiconductor data test abnormality warning system according to claim 4, characterized in that: The information acquisition module includes: An information request unit, used for requesting semiconductor data test information from the test terminal; Based on the semiconductor data test abnormal warning demand, send a semiconductor data test information request instruction to the test terminal, requesting the test terminal to transmit the semiconductor data test information; An information transmission unit, used for transmitting semiconductor data test information; After the test terminal receives the semiconductor data test information request instruction sent by the mobile terminal, the test terminal transmits the semiconductor data test information to the mobile terminal; An information acquisition unit, used for acquiring semiconductor data test information; The mobile terminal receives the semiconductor data test information transmitted by the test terminal, the mobile terminal acquires the semiconductor data test information, and determines the semiconductor data test information; The information processing module includes: An information retrieval unit, used for retrieving semiconductor data test information; Acquire semiconductor data test information, and search the semiconductor data test information based on a sequential search method; Filter out semiconductor data test information that is not valuable for semiconductor data test abnormality warning, and determine semiconductor data test information that is valuable for semiconductor data test abnormality warning; A feature extraction unit, used for extracting features from semiconductor data test information; Semiconductor data test information valuable for semiconductor data test abnormality warning is obtained, and features of the semiconductor data test information are extracted to determine semiconductor test feature data.

6. A semiconductor data test abnormality warning system according to claim 5, characterized in that: The data analysis module includes: A data storage unit, used to store pre-set semiconductor test standard data, providing a reference and guidance basis for semiconductor data testing; A data indexing unit, used for indexing pre-set semiconductor test standard data; Acquire semiconductor test characteristic data, and based on the semiconductor test characteristic data, index semiconductor test standard data matching the semiconductor test characteristic data from a data storage unit, and determine the indexed semiconductor test standard data; A data retrieval unit, used to retrieve the indexed semiconductor test standard data; Obtaining the indexed semiconductor test standard data, based on the semiconductor test standard data, and retrieving the semiconductor test standard data; A comparison and analysis unit, used for comparing and analyzing semiconductor test characteristic data; Acquire semiconductor test characteristic data and semiconductor test standard data, perform comparative analysis on the semiconductor test characteristic data based on the semiconductor test standard data, and determine corresponding semiconductor data test analysis results.

7. A semiconductor data test abnormality warning system according to claim 6, characterized in that: Comparative analysis unit, including: A standardization unit, used for performing data standardization on semiconductor test characteristic data according to data format characteristics of semiconductor test standard data to obtain target semiconductor test characteristic data; A data analysis unit, used to calculate the data importance P of the semiconductor test standard data based on the participation weight of the data type in the standard test result and according to the following formula; Where n represents the number of data types of semiconductor test standard data, N i represents the data volume of the i-th data type in the semiconductor test standard data, ℇ i represents the participation weight of the i-th data type in the semiconductor test standard data, θ Indicates the test difficulty of the standard test result, D j represents the amount of data of the jth data type in the standard test results, ℇ j represents the participation weight of the jth data type in the standard test result, and m represents the number of all data types in the standard test result; A comparison unit, used for calculating the comparison similarity S between the target semiconductor test feature data and the semiconductor test standard data based on the data importance of the semiconductor test standard data and according to the following formula; Among them, K0 represents the data characteristics of semiconductor test standard data, K a data features representing target semiconductor test feature data; The result determination unit is used to determine the corresponding semiconductor data test analysis result based on the comparison similarity between the target semiconductor test characteristic data and the semiconductor test standard data.

8. A semiconductor data test abnormality warning system according to claim 7, characterized in that: The abnormal warning module includes: A mining and analysis unit, used for mining and analyzing semiconductor data test analysis results; Obtain semiconductor data test analysis results, conduct in-depth mining and analysis on the semiconductor data test analysis results based on data mining technology, and determine the corresponding semiconductor data test abnormality warning method; An abnormal warning unit is used to issue an abnormal warning for semiconductor data test conditions; A semiconductor data test abnormality warning method is obtained, and an abnormality warning is performed on the semiconductor data test based on the semiconductor data test abnormality warning method.

Images