Search system, search method, and search program

Abstract

To easily obtain a data set related to a one factor experiment from a database.SOLUTION: The search system comprises at least one processor. The at least one processor is configured to access a database that stores a plurality of data records each indicating a correspondence between an identifier of a blending process and an element value in the blending process for each of a plurality of raw materials, select one of the plurality of data records as a reference record, and for each of one or more target records that are one or more data records different from the reference record among the plurality of data records, the number of raw materials whose element values are different from each other between the reference record and the object record is specified as a count value, and when there are one or more object records whose count value is 1 or less, the combination of the one or more object records and the reference record is extracted as a single factor experiment group.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] One aspect of the present disclosure relates to a search system, a search method, and a search program. [Background technology]

[0002] Systems using databases for material development have been known for some time. For example, Patent Document 1 describes a data management system for automatically and appropriately selecting data related to compositions. The data management system performs the following steps: acquiring first composition data indicating N compositions composed of at least one raw material among M candidate raw materials; acquiring reference composition data indicating a reference composition composed of at least one raw material among the M candidate raw materials; calculating the similarity between each row of an N × M matrix corresponding to the first composition data and an M-dimensional vector corresponding to the reference composition data; selecting, from the N rows of the N × M matrix, rows whose similarity to the M-dimensional vector satisfies a given criterion; generating second composition data composed of data records corresponding to the selected rows and having a number of dimensions smaller than M; and outputting the second composition data. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7302229 Summary of the Invention [Problem to be solved by the invention]

[0004] A mechanism for easily obtaining data sets related to single-factor experiments from databases is desirable. [Means for solving the problem]

[0005] A search system according to an aspect of the present disclosure includes at least one processor that accesses a database storing a plurality of data records indicating correspondence between a blending process identifier and an element value in the blending process for each of a plurality of raw materials, selects one of the plurality of data records as a base record, and, for each of one or more target records that are one or more data records different from the base record, identifies, as a count value, the number of raw materials whose element values ​​differ between the base record and the target record, and, if there are one or more target records whose count value is 1 or less, extracts combinations of the one or more target records and the base record as a one-factor experimental group.

[0006] In this aspect, a certain data record is selected as a reference record, target records that have element values ​​different from the reference record for a single raw material are identified, and the combination of the reference record and the target record is extracted as a single-factor experiment group. By this series of processes, a data set related to a single-factor experiment can be easily obtained from a database. [Effects of the Invention]

[0007] According to one aspect of the present disclosure, data sets relating to single-factor experiments are readily available from databases. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of a functional configuration of a search system. [Figure 2] FIG. 10 is a diagram showing an example of combination data. [Figure 3] 10 is a flowchart illustrating an example of a full search by the search system. [Figure 4] FIG. 10 is a diagram showing an example of extracting a one-factor experimental group by exhaustive search. [Figure 5] 10 is a flowchart illustrating another example of processing by the search system. DETAILED DESCRIPTION OF THE INVENTION

[0009] Various examples of the present disclosure will be described in detail below with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements are designated by the same reference numerals, and redundant description will be omitted.

[0010] [System Overview] The search system according to the present disclosure is a computer system that extracts a dataset related to a single-factor experiment from a database that stores formulation data related to formulation processing. In one example, by using the search system, a user can easily obtain a dataset from a database that accumulates a huge amount of formulation data.

[0011] Blending refers to the process of obtaining a composition from two or more raw materials. Examples of compositions include composite materials and mixtures. The blending method can be selected appropriately depending on the type of composition.

[0012] A single-factor experiment is a method of repeating a treatment (experiment) while changing the level of only one factor. Generally, in the treatment of a composition, a single-factor experiment is conducted for each of several factors (raw materials) for the same base composition, which is a combination of two or more specific raw materials. The influence of the factors on the properties of the composition is then evaluated by comparing the results of each treatment using a specific property as a standard.

[0013] Each formulation process produces a composition from two or more selected raw materials. To obtain a composition with the desired functionality, each worker repeatedly selects two or more raw materials from a vast number of candidate raw materials and performs the formulation process. The results of each process are stored in a database as a data record of formulation data. The number of raw materials recorded in the database increases with each new raw material used, potentially reaching 1,000 or more. Furthermore, because formulations are performed based on various perspectives or considerations, the database can become a collection of data sets with little correlation to each other. Due to these factors, the accumulated formulation data as a whole becomes sparse data, with most element values ​​related to the raw materials being zero or null. This makes it extremely difficult to identify a single-factor experimental group from the database. However, by using a search system, users can easily obtain a dataset representing a single-factor experimental group. Users can refer to this dataset, which can be considered knowledge gained from previous processes, to further consider or experiment with formulation processes.

[0014] [System Configuration] A search system is made up of one or more computers. When multiple computers are used, these computers are connected via a communication network such as the Internet or an intranet to logically construct a single search system.

[0015] The computers that make up a search system generally have a processor, a storage device (memory), and a communication interface as hardware devices. The processor is, for example, a CPU, and the storage device is composed of flash memory, a hard disk, etc. The communication interface is composed of a network card, a wireless communication module, etc. Each functional module of the search system is realized by the processor executing a program stored in the storage device.

[0016] A search program that causes a computer to function as a search system includes program code for implementing each functional module of the search system. This search program may be provided in a state recorded on a non-transitory recording medium such as a CD-ROM, DVD-ROM, or semiconductor memory. Alternatively, the search program may be provided via a communication network as a data signal superimposed on a carrier wave. The provided search program is recorded, for example, in a storage device.

[0017] 1 is a diagram showing the functional configuration of a search system 10 according to an example. In one example, the search system 10 is connected to a database 20 and a user terminal 30 via a communication network. The communication network is typically constructed using the Internet, an intranet, or a combination thereof. The communication network may be constructed using a wired network, a wireless network, or a combination thereof.

[0018] The database 20 is a device that stores blending data related to blending processes. The blending data is composed of a plurality of data records (e.g., a large number of data records). Each data record indicates a process ID, which is an identifier of the blending process, and a correspondence between element values ​​in the blending process for each of a plurality of raw materials. The blending data may include other data items in addition to the process ID and the raw material list. For example, the blending data may include at least one of the attributes of the composition and the conditions or environment of the work as data items.

[0019] FIG. 2 is a diagram showing an example of the composition data in the database 20. As shown in FIG. 2, in this disclosure, a collection of multiple raw materials (candidate raw materials) is also referred to as an "raw material list." As mentioned above, the number of raw materials that make up the raw material list can be enormous, for example, 1000 or more. In FIG. 2, each element value is represented by r 11 ,r 12Each element value may be expressed by the amount or mixing ratio of the raw material used in the blending process, or by other types of quantities. The element values ​​of raw materials not used in the blending process are set to, for example, 0 or null.

[0020] The user terminal 30 is a computer used by a user of the search system 10. The user terminal 30 may be any of various computers, such as a personal computer, a workstation, a tablet terminal, a smartphone, or a wearable terminal.

[0021] The search system 10 includes a processor 101 that functions as a search unit 11. The search unit 11 is a functional module that accesses the database 20 based on instructions from a user terminal 30, extracts single-factor experimental groups, and provides the extraction results (search results) to the user terminal 30. A single-factor experimental group refers to a combination of one specific blending process and one or more other blending processes that are in a single-factor experimental relationship with the specific blending process.

[0022] [System Operation] As an example of a search method according to the present disclosure, a full search of the database 20 by the search system 10 will be described with reference to Fig. 3. Fig. 3 is a flowchart showing an example of the full search as a processing flow S1.

[0023] In step S101, the search unit 11 receives a user instruction from the user terminal 30. This user instruction is a data signal that causes the search system 10 to execute a full search of the database 20. The user performs an operation for the full search on the user terminal 30. In response to the user operation, the user terminal 30 generates a user instruction and transmits the user instruction to the search system 10. The search unit 11 receives the user instruction.

[0024] In step S102, the search unit 11 initializes a variable i to 1. This variable i serves to identify a reference record used as a reference for extracting a one-factor experimental group.

[0025] In step S103, the search unit 11 accesses the database 20 and retrieves one data record R i is selected as the base record.

[0026] In step S104, the search unit 11 initializes a variable j to 1. This variable j is the reference record R i It serves to identify the target record, which is the data record that is compared with the target record.

[0027] In step S105, the search unit 11 accesses the database 20 and retrieves one data record R j Select as the target record.

[0028] In step S106, the search unit 11 searches the reference record R i and target record R j The number of ingredients whose element values ​​differ between the two data records is identified as a count value t. Assuming that each data record is a row and each ingredient in the ingredient list is a column, the search unit 11 determines for each column whether the element values ​​differ between the two data records. The search unit 11 then identifies the number of columns whose element values ​​differ between the two data records as a count value t.

[0029] In step S107, the search unit 11 determines whether the count value t is 1 or less. If the count value t is 1 or less, the search unit 11 searches the reference record R i The list L corresponding to i to the target record R j Add a list L i The initial state of the list L is an empty list, and the list L is added by the search unit 11. i The target record can be recorded in the target record R j List L i As a process to add to the target record R j the target record R jFor example, the value of variable j or a process ID may be added.

[0030] As shown in step S108, the search unit 11 searches for one reference record R i All data records of the blending data are processed as target records.

[0031] If an unprocessed target record exists (NO in step S108), the process proceeds to step S109, where the search unit 11 increments the variable j by 1. Then, the process returns to step S105. In the repeated step S105, the search unit 11 searches for the next data record R j The search unit 11 selects the target record R j The processes of steps S106 and S107 are executed for the above.

[0032] All data records of the compounding data are included in the target record R. j If the result of the process is "YES" in step S108, the process proceeds to step S110. i Determine whether the number of data records in list L is 2 or more. i is the reference record R i The set of a specific combination treatment indicated by the formula L and other combination treatments that are in a one-factor experimental relationship with the specific combination treatment is shown as a one-factor experimental group. i contains three or more data records, the base record R i It should be noted that two or more records other than those listed above are not necessarily related to one-factor experiments. If the number of data records is two or more, the search unit 11 creates a list L to be used as the extraction result (search result). all Listed on L i If the number of data records is less than 2, the list L is added without performing the addition process. i is rejected.

[0033] If i=j, that is, if the target record is the same as the base record, the count value t becomes 0 in step S106. i List L i Therefore, the list L i If the number of data records in is 2 or more, the base record R i One or more target records R in a one-factor experiment j means that there exists a list L i The number of data records in is 1, which means that the base record R i This means that there is no single-factor experimental group corresponding to

[0034] In this way, the search unit 11 selects the reference record R from among the plurality of data records. i One or more target records R that are one or more data records that are different from j For each of these, the reference record R i and the target record R j The search unit 11 then identifies the number of raw materials whose element values ​​are different between the raw materials R and the raw materials R, as a count value t. j (where i≠j) exists, then the one or more target records R j and the reference record R i The combination of is extracted as a single-factor experimental group.

[0035] As shown in step S111, the search unit 11 processes all data records of the combination data as reference records.

[0036] If there is a data record that has not been processed as a base record (NO in step S111), the process proceeds to step S112, where the search unit 11 increments the variable i by 1. Then, the process returns to step S103. In the repeated step S103, the search unit 11 searches for the next data record R i The search unit 11 selects the reference record R iThe process of steps S104 to S110 is executed for each of the above.

[0037] All data records of the formulation data are stored as the base record R i If the result of the process is "YES" in step S111, the process proceeds to step S113. all After removing duplicates in the list L all is transmitted to the user terminal 30. In some cases, multiple lists L, each of which is designated as a single-factor experimental group, are transmitted. i Between two or more lists L with the same combination of data records i There may be such a list L i and remove the duplicates from the list L all The search unit 11 sends the list L to the user terminal 30. The list L is edited to clearly show the raw materials and element values ​​used in each blending process. all The user terminal 30 may transmit the list L all , which allows the user to obtain one or more single-factor experimental groups.

[0038] In this way, the search unit 11 extracts a single-factor experimental group when there are one or more target records whose count value is 1 or less for each of two or more base records while changing the base record among the multiple data records. The search unit 11 may also remove overlapping single-factor experimental groups among two or more single-factor experimental groups, and obtain one or more single-factor experimental groups as the extraction result.

[0039] Figure 4 shows an example of extracting a one-factor experimental group by full search. In this example, for simplicity, the raw material list consists of five raw materials Ma, Mb, Mc, Md, and Me, and the composition data in database 20 consists of six data records. Below, individual data records will be distinguished by their process IDs.

[0040] The search unit 11 selects the data record "P1001" as the base record (step S103). Then, for each of the six target records, the search unit 11 specifies the number of ingredients whose element values ​​are different between the base record "P1001" and the target record as a count value t, and adds the target records whose count value t is 1 or less to the list L1 (repeated steps S105 to S109). As shown in FIG. 4, the data records whose count value t is 1 or less in relation to the data record "P1001" are the data record "P1001", the data record "P1002", and the data record "P1005". The one-factor experiment group indicated by the list L1 indicates the data records that are in a one-factor experiment relationship with the data record "P1001". As described above, the list L i In this example, two or more records other than the reference record are not necessarily in a one-factor experiment relationship. In the example of list L1 in Figure 4, data record "P1002" and data record "P1005" are not in a one-factor experiment relationship.

[0041] Since the number of data records is two or more, the search unit 11 searches the list L1 for the list L all (step S110).

[0042] Next, the search unit 11 selects each of the remaining five data records as a reference record and creates a list L i (i=2~6) and generate a list L with 2 or more data records. i List L all (Steps S103 to S112 are repeated). Since the list L4 corresponding to the data record "P1004" and the list L6 corresponding to the data record "P1006" both contain only the reference record, the search unit 11 rejects these two lists. Finally, the search unit 11 adds the list L all Since the lists L1, L2, L3, and L5 do not overlap with each other, the search unit 11 adds the lists L1, L2, L3, and L5 to the list L1 without removing the overlaps. all is transmitted to the user terminal 30 (step S113).

[0043] As another example of a search method according to the present disclosure, a search using a specific data record as a reference record will be described with reference to Fig. 5. Fig. 5 is a flowchart showing this example of a search as process flow S2.

[0044] In step S201, the search unit 11 receives search conditions as user instructions from the user terminal 30. The user operates the user terminal 30 to specify search conditions to be used as criteria for searching for a desired single-factor experimental group. The search conditions may be set based on information about a single blending process, for example, the combination of ingredients in the blending process and the element values ​​of each ingredient. The user terminal 30 generates search conditions in response to the user's operation and transmits an instruction signal indicating the search conditions to the search system 10. The search unit 11 receives the search conditions.

[0045] In step S202, the search unit 11 accesses the database 20 and retrieves the data records R that satisfy the search conditions. d is selected as the base record.

[0046] In step S203, the search unit 11 initializes a variable j to 1. This variable j is the reference record R d It serves to identify the target record, which is the data record that is compared with the target record.

[0047] In step S204, the search unit 11 accesses the database 20 and retrieves one data record R j Select as the target record.

[0048] In step S205, the search unit 11 searches the reference record R d and target record R j The number of ingredients whose element values ​​differ between the ingredients is identified as a count value t. This identification method is the same as in step S106.

[0049] In step S206, the search unit 11 determines whether the count value t is 1 or less. If the count value t is 1 or less, the search unit 11 searches the reference record R d The list L corresponding to d to the target record R j Add a list L d The initial state of the list L is an empty list, and the list L is added by the search unit 11. d As in step S107, the search unit 11 searches for the target record R j the target record R j For example, the value of variable j or a process ID may be added.

[0050] As shown in step S207, the search unit 11 searches the reference record R d All data records of the blending data are processed as target records.

[0051] If an unprocessed target record exists (NO in step S207), the process proceeds to step S208, where the search unit 11 increments the variable j by 1. Then, the process returns to step S204. In the repeated step S204, the search unit 11 searches for the next data record R j The search unit 11 selects the target record R j The processes of steps S205 and S206 are executed for the

[0052] All data records of the compounding data are included in the target record R. j If the result of the process is "YES" in step S207, the process proceeds to step S209. d Determine whether the number of data records in list L is 2 or more. d is the reference record R d The set of a specific combination treatment indicated by the formula L and other combination treatments that are in a one-factor experimental relationship with the specific combination treatment is shown as a one-factor experimental group. d contains three or more data records, the base record R dIt should be noted that two or more records other than those listed above are not necessarily related to one-factor experiments. d The search unit 11 sends the extracted results (search results) to the user terminal 30. The search unit 11 generates a list L that is edited to clearly show the raw materials and element values ​​used in each blending process. d The user terminal 30 may transmit the list L d The list L is received and displayed, allowing the user to obtain a single-factor experimental group corresponding to the search criteria. d If the number of data records in is less than two, the search unit 11 may send a message to the user terminal 30 indicating that no single-factor experimental group corresponding to the search conditions was extracted. The user terminal 30 receives and displays the message.

[0053] In this way, in the process flow S2, the search unit 11 also searches for the reference record R d One or more target records R that are one or more data records that are different from j For each of these, the reference record R d and the target record R j The search unit 11 then identifies the number of raw materials whose element values ​​are different between the raw materials R and the raw materials R, as a count value t. j (where i≠j) exists, then the one or more target records R j and the reference record R d The combination of is extracted as a single-factor experimental group.

[0054] As a modification of the process flow S2, two or more data records R that satisfy the search conditions received from the user terminal 30 are retrieved. d If there is a data record R, the search unit 11 d Then, the search unit 11 executes steps S202 to S209 for each data record R d The user terminal 30 transmits the extraction results (search results) for the list L. dIn this example, the search unit 11 also changes the base record among the multiple data records, and extracts a one-factor experimental group when there are one or more target records with a count value of 1 or less for each of two or more base records.

[0055] [Variations] The technology according to the present disclosure has been described in detail above based on various examples. However, the present disclosure is not limited to the above examples. The technology according to the present disclosure can be modified in various ways without departing from the spirit of the present disclosure.

[0056] In the above example, an algorithm is used that compares the base record and the target record even when they are the same data record (i.e., when i = j). However, the algorithm that realizes the search system is not limited to this example. For example, an algorithm that omits comparing the base record and the target record when they are the same may be used. In this modified example, the search system also identifies, for each of one or more target records that are one or more data records different from the base record among multiple data records, the number of raw materials whose element values ​​differ between the base record and the target record as a count value. Then, if there are one or more target records with a count value of 1 or less, the search system extracts combinations of the one or more target records and the base record as a one-factor experimental group.

[0057] In the above example, the search system 10 acts as a server in a client-server system. As another example, the functions of the search system 10 and the database 20 may be implemented in a stand-alone computer. Alternatively, the search system may be implemented in a user terminal that can access the database 20 via a communication network.

[0058] The processing steps of the method executed by at least one processor are not limited to the above examples. For example, some of the above steps may be omitted, or the steps may be executed in a different order. Furthermore, any two or more of the above steps may be combined, or some of the steps may be modified or deleted. Alternatively, other steps may be executed in addition to the above steps.

[0059] In the present disclosure, when comparing the magnitude of two numerical values, either of the two criteria "greater than or equal to" and "greater than" may be used, or either of the two criteria "less than or equal to" and "less than" may be used.

[0060] In this disclosure, the expression "at least one processor executes a first process, executes a second process, ... executes an nth process" or a corresponding expression indicates a concept including a case where the entity executing the n processes from the first process to the nth process, i.e., the processor, changes midway through. In other words, this expression indicates a concept including both a case where all n processes are executed by the same processor and a case where the processor changes among the n processes according to an arbitrary policy.

[0061] [Note] As can be seen from the various examples above, the present disclosure includes the following aspects. (Appendix 1) at least one processor; the at least one processor: accessing a database storing a plurality of data records indicating correspondence between a blending process identifier and an element value in the blending process for each of a plurality of raw materials; selecting one of the plurality of data records as a reference record; For each of one or more target records, which are one or more data records different from the reference record among the plurality of data records, the number of the raw materials whose element values ​​differ between the reference record and the target record is identified as a count value; If there are one or more target records whose count value is 1 or less, extracting a combination of the one or more target records and the reference record as a one-factor experimental group. Search system. (Appendix 2) the at least one processor extracts the one-factor experimental group when there are one or more target records whose count value is 1 or less for each of two or more of the base records while changing the base record among the plurality of data records; 1. The search system described in Appendix 1. (Appendix 3) the at least one processor removes overlapping one-factor experimental groups among two or more of the one-factor experimental groups to obtain one or more one-factor experimental groups as an extraction result; The search system described in Appendix 2. (Appendix 4) 1. A search method performed by a search system comprising at least one processor, comprising: accessing a database storing a plurality of data records indicating a correspondence between a blending process identifier and an element value in the blending process for each of a plurality of raw materials; selecting one of the plurality of data records as a base record; a step of identifying, for each of one or more target records which are one or more data records different from the reference record among the plurality of data records, the number of the raw materials whose element values ​​are different between the reference record and the target record as a count value; When there are one or more target records whose count value is 1 or less, extracting a combination of the one or more target records and the reference record as a one-factor experimental group; Search methods including. (Appendix 5) accessing a database storing a plurality of data records indicating a correspondence between a blending process identifier and an element value in the blending process for each of a plurality of raw materials; selecting one of the plurality of data records as a base record; a step of identifying, for each of one or more target records which are one or more data records different from the reference record among the plurality of data records, the number of the raw materials whose element values ​​are different between the reference record and the target record as a count value; When there are one or more target records whose count value is 1 or less, extracting a combination of the one or more target records and the reference record as a one-factor experimental group; A search program that causes a computer to execute the following.

[0062] According to Supplements 1, 4, and 5, a certain data record is selected as a base record, target records that have element values ​​different from the base record for a single raw material are identified, and the combination of the base record and the target record is extracted as a single-factor experiment group. Through this series of processes, a dataset related to a single-factor experiment can be easily obtained from a database.

[0063] According to Appendix 2, since extraction of a single-factor experimental group is attempted for each of two or more reference records, it becomes possible to comprehensively extract single-factor experimental groups. For example, it becomes possible to perform a comprehensive search of the entire database.

[0064] According to Appendix 3, since the overlap of the single-factor experimental group is eliminated, extraction results that are easy for users to recognize or process can be obtained, which can lead to improved convenience of the search system. [Explanation of symbols]

[0065] 10...search system, 11...search unit, 20...database, 30...user terminal.

Claims

1. at least one processor; the at least one processor: accessing a database storing a plurality of data records indicating a correspondence between a blending process identifier and an element value in the blending process for each of a plurality of raw materials; selecting one of the plurality of data records as a reference record; For each of one or more target records, which are one or more data records different from the reference record among the plurality of data records, the number of the raw materials whose element values ​​differ between the reference record and the target record is identified as a count value; If there are one or more target records whose count value is 1 or less, a combination of the one or more target records and the reference record is extracted as a one-factor experimental group. Search system.

2. the at least one processor extracts the one-factor experimental group when there are one or more target records whose count value is 1 or less for each of two or more of the base records while changing the base record among the plurality of data records; The search system according to claim 1 .

3. the at least one processor removes overlapping one-factor experimental groups among two or more of the one-factor experimental groups to obtain one or more one-factor experimental groups as an extraction result; The search system according to claim 2 .

4. 1. A search method performed by a search system comprising at least one processor, comprising: accessing a database storing a plurality of data records indicating a correspondence between a blending process identifier and an element value in the blending process for each of a plurality of raw materials; selecting one of the plurality of data records as a base record; a step of identifying, for each of one or more target records which are one or more data records different from the reference record among the plurality of data records, the number of the raw materials whose element values ​​are different between the reference record and the target record as a count value; extracting a combination of one or more target records and the reference record as a one-factor experimental group when there are one or more target records whose count value is 1 or less; Search methods including.

5. accessing a database storing a plurality of data records indicating a correspondence between a blending process identifier and an element value in the blending process for each of a plurality of raw materials; selecting one of the plurality of data records as a base record; a step of identifying, for each of one or more target records which are one or more data records different from the reference record among the plurality of data records, the number of the raw materials whose element values ​​are different between the reference record and the target record as a count value; extracting a combination of one or more target records and the reference record as a one-factor experimental group when there are one or more target records whose count value is 1 or less; A search program that causes a computer to execute the following.

Images