WO2020082992A1 - Method for learning word by using historical spelling experience, apparatus and electronic device - Google Patents

Abstract

A method for learning a word by using historical spelling experience, an apparatus and an electronic device, the method comprising: acquiring at least one spelling partition combination of a target word, the spelling partition combination comprising at least one spelling partition (101); obtaining a partition spelling result containing historical spelling experience of the target word on the basis of the at least one spelling partition combination and an associated historical spelling record of the spelling partition in the spelling partition combination (102). By means of the described method, a partition spelling result containing historical spelling experience of a target word is obtained by means of acquiring at least one spelling partition combination of the target word and on the basis of the at least one spelling partition combination and an associated historical spelling record of the spelling partition in the spelling partition combination. Thus, a partition spelling prompt containing correct historical spelling experience may be provided when a learner learns an unfamiliar word so as to prevent the learner from using a historical spelling memory by mistake.

Description

Word learning method, device and electronic equipment borrowing historical spelling experience Technical field

Embodiments of the present invention relate to the field of electronic-assisted teaching technology, and in particular, to a word learning method, device, and electronic device that borrow historical spelling experience.

Background technique

In the system of "letter spelling text", we can rely on the intuitive way to observe the letter string writing sequence of the word itself, and combine the individual pronunciations represented in it into syllable pronunciations, and then merge into the overall pronunciation of the word. This process of recognition is called "natural spelling".

"Natural spelling" is spelled directly from the alphabet string, which has the disadvantage of uncertainty in pronunciation in many languages. Because the letters and bytes of each word in the "natural spelling" can have more than one pronunciation under the combination of different letter strings before and after. Due to the uncertainty of pronunciation, learners can easily use historical spelling memory in the pronunciation of strange words. Because foreign language natural spelling learners lack the environmental advantage of using the language as the mother tongue, the phenomenon of erroneous borrowing of associated spelling memory is particularly serious and common among such learners. For example: "different" and "rent" both contain the same letter string "rent", but the letter string "rent" is pronounced differently in these two words, the word "rent" is pronounced / rent / in "different" Is pronounced in

When the learner encounters the strange word "different", it is easy to borrow the pronunciation of the word "rent" that he is familiar with and copy it into "different", thereby causing a wrong pronunciation.

Summary of the invention

The technical problem mainly solved by the embodiments of the present invention is to provide a word learning method, device and electronic equipment that borrow historical spelling experience. When a learner learns a strange word, it can provide a section spelling prompt including correct historical spelling experience. Avoid learners from borrowing historical spelling memory by mistake.

To solve the above technical problem, in a first aspect, an embodiment of the present invention provides a word learning method borrowing historical spelling experience, the method includes:

Obtaining at least one spelling partition combination of the target word, where the spelling partition combination includes at least one spelling partition;

Based on the at least one spelling partition combination and the associated historical spelling records of the spelling partitions in the spelling partition combination, a partition spelling result containing historical spelling experience of the target word is obtained.

In a second aspect, an embodiment of the present invention also provides a word learning device that borrows historical spelling experience. The device includes:

A spelling partition combination acquisition module, configured to acquire at least one spelling partition combination of the target word, and the spelling partition combination includes at least one spelling partition;

The spelling result acquisition module is used to obtain the segmentation spelling of the target word containing the historical spelling experience based on the at least one spelling division combination and the associated historical spelling records of the spelling division in the spelling division combination result.

In a third aspect, an embodiment of the present invention also provides an electronic device, the electronic device includes:

At least one processor; and,

A memory communicatively connected to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above method.

According to a fourth aspect, an embodiment of the present invention further provides a non-volatile computer-readable storage medium, where the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by an electronic device, Causing the electronic device to perform the method described above.

The embodiment of the present invention obtains the target word's Partition spelling results including historical spelling experience. When learners are learning strange words, they can provide section spelling tips including correct historical spelling experience to avoid learners from borrowing historical spelling memory by mistake.

BRIEF DESCRIPTION

In order to more clearly explain the technical solutions of the embodiments of the present invention, the drawings required in the embodiments of the present invention will be briefly described below. Obviously, the drawings described below are only some embodiments of the present invention. For a person of ordinary skill in the art, without paying any creative work, other drawings can also be obtained based on these drawings.

1 is a schematic diagram of an application scenario of a word learning method and device borrowing historical spelling experience of the present invention;

2 is a flowchart of an embodiment of a word learning method borrowing historical spelling experience of the present invention;

3 is a schematic structural diagram of an embodiment of a word learning device borrowing historical spelling experience of the present invention;

4 is a schematic structural diagram of an embodiment of a word learning device borrowing historical spelling experience of the present invention;

5 is a schematic diagram of a hardware structure of an electronic device provided by an embodiment of the present invention.

detailed description

In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

It should be noted that all the expressions using "first" and "second" in the embodiments of the present invention are for distinguishing two entities with the same name but not the same or different parameters. See "first" and "second" For convenience of expression only, it should not be understood as a limitation of the embodiments of the present invention, and subsequent embodiments will not describe them one by one.

The word learning method and device for borrowing historical spelling experience provided by the embodiments of the present invention are applicable to the application scenario shown in FIG. 1. In the application scenario shown in FIG. 1, the user 10 and the electronic device 20 are included. Among them, the electronic device 20 is, for example, a learning machine, a smart phone, a personal computer, a tablet computer, a robot, a wearable smart device, a cloud server, and the like. The user 10 can use the electronic device 20 to learn a foreign language that belongs to alphabetic spelling.

In the "letter spelling" system, the "natural spelling" method can avoid memorizing pronunciation codes ("pronunciation codes" refer to codes that can characterize a certain pronunciation, such as phonetic strings or other specific codes corresponding to related pronunciations) , So it has been widely used. However, the letters and bytes of each word in the "natural spelling" can have more than one pronunciation under the combination of different strings of letters before and after. When the user 10 learns using the "natural spelling" method, due to the uncertainty of pronunciation, the user 10 easily applies the historical spelling memory to the pronunciation of strange words. For example: "different" and "rent" both contain the same letter string "rent". When user 10 encounters the strange word "different", it is easy to borrow the pronunciation of the familiar word "rent" and copy it to "different" This can cause incorrect pronunciation.

When the user 10 uses the electronic device 20 to learn a foreign language, if the electronic device 20 can prompt the user 10 of the correct historical spelling memory of the strange word (hereinafter referred to as the target word) he learns, the user 10 can be effectively prevented from borrowing the historical spelling by mistake Read memory. Taking the above example as an example, suppose the user 10 has learned the words "difficult" and "reference". When the user 10 learns the target word "different", if the electronic device 20 can spell out the division of "different", for example, it is divided into " dif-feren-t ", and shows the historical spelling words associated with each spelling partition (that is, the words that the user has learned that are associated with the spelling partition), then it can well prevent the user 10 from borrowing the wrong historical memory . For example, the electronic device 20 may prompt the user 10 as follows: The spelling method of "different" is: the first tone is "dif" in "difficult", and the second tone is "feren" in "reference" , Followed by the ending "t".

Since each word may have multiple partitioning methods, each partition may be associated with multiple historical spelling words. In some embodiments, the electronic device 20 may prompt the user of the spelling partitions in various combinations of spelling partitions of the target word, as well as the respective historical spelling words of each spelling partition. In other embodiments, in order to be more targeted and make the prompt result simple and clear, the electronic device 20 may select a preferred spelling from multiple spelling partitions and multiple associated historical spelling words according to preset filtering rules Read the partition and its associated history spelling words, and prompt the user 10 according to the preferred spelling partition and its associated history spelling words.

Wherein, the "letter spelling text" is, for example, English, German, French, Greek, Italian, Portuguese, etc. Therefore, in the embodiment of the present invention, the "target word" may be a word in any of the above languages. In order to facilitate explanation of the inventive concept of the present invention, in the embodiments of the present invention, the target word is mainly English or American English words as an example for detailed description.

FIG. 2 is a schematic flowchart of a word learning method for borrowing historical spelling experience provided by an embodiment of the present invention. The method may be performed by the electronic device 20 in FIG. 1, as shown in FIG. 2, the method includes:

101: Acquire at least one spelling partition combination of a target word, where the spelling partition combination includes at least one spelling partition.

Among them, the spelling partition combination represents a way of segmenting the spelling partition. The spelling partition combination is composed of several spelling partitions. Each spelling partition can be, for example, a string of letters, a string of pronunciation codes, or a group Corresponding letter string and pronunciation code. For example, the word different can include a combination of spelling partitions

Wait, where,

t / t /, fe / f / and

Partition for spelling.

Wherein at least one spelling partition combination of the target word is obtained, a sub-library of possible spelling partition combinations of words may be set in the database of the electronic device 20 in advance, and the sub-library of possible spelling partition combinations of words stores multiple words and each word Corresponding combinations of various possible spelling partitions (in other embodiments, the database may also be set on other servers, and the electronic device 20 accesses the databases of other servers through the network). By searching for the target word in the sub-base of possible spelling partition combinations of words, each possible spelling partition combination of the target word can be obtained. A plurality of words are stored in the sub-library of possible spelling partition combinations of the words, and each word corresponds to several possibility partition combinations, and several spelling partitions included in each possibility partition combination. In some of these embodiments, the word possibility partition combination sub-library may also store associated words corresponding to the spelling partitions of each word and / or associated spelling partitions, and the probability of correct structural pronunciation corresponding to each spelling partition . Please refer to Table 1. Table 1 exemplarily shows one of the structural forms of data storage for each word in the sub-base of possible combination of possible spelling partitions.

The probability of correct structural pronunciation is characterized by the ratio of words pronounced with the pronunciation code corresponding to the spelling partition in each word including the letter string corresponding to the spelling partition. For example, spelling partition

The letter string "dif" is sent in every word containing "dif"

Sound, spelling partition

The probability of correct structural pronunciation is 100%. Spelling partition

The letter string "ferent" is sent in most words containing "ferent"

Sound, can be calculated by including the letter string "ferent" and pronounced

The proportion of words with sounds in the words containing the letter string "ferent" to obtain the spelling partition

The probability of correct pronunciation of the structured pronunciation is 99.88% (the pronunciation of "ferent" here is exemplified by the annotation of the Oxford dictionary. When the method of the embodiment of the present invention is applied in other countries or regions, the spelling can be based on actual application (The probability of the structural pronunciation of the partition is calculated correctly).

The related words corresponding to the spelling partition refer to the words including the spelling partition, that is, the related words include both the letter string corresponding to the spelling partition, and the pronunciation of the letter string in the related word is the pronunciation corresponding to the spelling partition The code is consistent. The associated spelling partition corresponding to the spelling partition is the spelling partition including the spelling partition. The spelling partition combinations, spelling partitions, structural pronunciation correct probability, and related words corresponding to the above-mentioned words can be preset in the sub-library of possible word spelling partition combinations.

Table 1

Note 1: In this embodiment, the British Standard (British, Received, Pronunciation, Accent) is used, so the ending of the letter at the end of the word is basically unchanged.

The above is only a method for obtaining possible combinations of possible spelling partitions of the target word. In other embodiments, the sub-library of possible combination of spelling partitions of the word may not be established in advance, but the preset Library to get every possible spelling partition combination of the target word.

Table 2

A plurality of spelling partitions are pre-stored in the spelling partition data sub-library, please refer to Table 2, Table 2 exemplarily shows a data structure form of the spelling partitions stored in the spelling partition data sub-library. Since different spelling partitions may have the same letter string, in order to facilitate distinction, and also to facilitate recording and querying, a unique spelling code can be configured for each spelling partition, for example, the spelling code P1A in Table 2 represents spelling Read partition

The spelling code P1B stands for if and

The spelling code can characterize the correspondence between the letter string and the pronunciation code. In some of these embodiments, the spelling partition data sub-library may also store associated words and / or associated spelling partitions corresponding to each spelling partition, and the probability of correct structural pronunciation. Therefore, according to the spelling partition data sub-library, the correspondence between the spelling partition (spelling code), the letter string, the pronunciation code, and the word containing the spelling partition can be obtained.

Specifically, each possible spelling partition combination of the target word is obtained based on the spelling partition data sub-base. You can find each spelling partition associated with the target word in the spelling partition data sub-base, and then perform Combination to obtain each possible spelling partition combination of the target word. Still taking "different" as an example, based on the spelling partition data sub-base (please refer to Table 2), find out the spelling partitions with the associated word d i f f e r e n t (for example

Etc.), and then combine the spelling partitions (the combination condition is that the letter string and pronunciation code after the combination of each spelling partition must be different and the corresponding whole word pronunciation code

). Among them, for the adjacent spelling partitions in the spelling partition combination, if the pronunciation code at the end of the previous spelling partition is the same as the first pronunciation code of the following spelling partition, only one of the pronunciation codes can be provided.

For example, for spelling partition combinations

Medium, due to spelling partition

Last letter r and spelling partition in

The first letter r in is the shared phoneme, and the shared phoneme only provides one r, then the letter string formed by the combination of the two spelling partitions is still d i f f e r e n t. Spelling partition combination

Can be used as a possible spelling partition combination of different.

In other embodiments of the present invention, the best spelling partition combination of the target word may be obtained first, and then the spelling partitions in the best spelling partition combination may be combined and / or subdivided to obtain At least one spelling partition combination of the target word.

The best spelling partition combination of the target word can be obtained based on the preset best spelling partition combination data sub-base. Among them, the best spelling partition combination data sub-bank stores multiple words, the best spelling partition combination of each word, and the spelling partition included in each spelling partition combination. Please refer to Table 3, which exemplarily shows a structure form for storing each word data in the sub-base of the best spelling partition combination data. In other embodiments, the best spelling partition combination data sub-library may also store associated words and / or associated spelling partitions of each spelling partition. The best spelling partition combination of a word is a spelling partition combination with a relatively high probability that the structural pronunciation of each spelling partition is correct among the possible spelling partition combinations of the word. For example, it may be a combination of spelling partitions with the highest probability of correct structural pronunciation among the possible combinations of spelling partitions of words. The comprehensive structural pronunciation correct probability can be obtained by using a mathematical average algorithm to calculate the structural pronunciation correct probability of each spelling partition in the spelling partition combination. Wherein, the mathematical average algorithm may include, but is not limited to: total mathematical average algorithm, intermediate average algorithm, mode average algorithm, RMS average algorithm, and so on.

By searching the target word in the sub-base of the best spelling partition combination data, the best spelling partition combination of the target word can be obtained. Then combining and / or subdividing the best spelling partition combination to obtain at least one spelling partition combination of the target word.

table 3

102: Based on the at least one spelling partition combination and the associated historical spelling records of the spelling partitions in the spelling partition combination, obtain a partition spelling result of the target word that includes historical spelling experience.

Wherein, the division spelling result containing the historical spelling experience may be a spelling division of the target word, and a spelling prompt for each spelling division according to the user 10's historical spelling experience of the spelling division. For example, for each spelling section, the related words in the spelling section and / or the related words and / or related spelling sections in which the user 10 is more familiar in the related spelling section may be prompted (the related words will be described below as an example). The related words that the user 10 is familiar with, that is, the related history spelling words are, for example, related words that the user 10 has learned and passed the assessment. The concept of related history spelling words may be defined on the electronic device 20 in advance. In some embodiments, the related history spelling words of the spelling section include related words that the user 10 has learned and evaluated related words. Due to different target words, the spelling partition in the spelling partition result may be a single spelling partition, or may include at least two spelling partitions.

When the user 10 uses the electronic device 20 to learn, the electronic device 20 can record the learning situation of the user 10 for each word, for example, the number of times the user 10 learns the word, the number of assessment passes, whether the pronunciation method is whole reading or spelling. Among them, the whole reading means that the pronunciation of the pronunciation of the word pronunciation is not required to be read out by the spelling partition of the word as a whole, and the pronunciation way of pronunciation of the word pronunciation is to be read out in a segmented manner by means of the spelling partition of the word. In practical applications, the electronic device 20 can recognize whether the pronunciation mode is full reading or spelling according to the pronunciation of the user 10. In other embodiments that cannot recognize the pronunciation mode, the user 10 can also input the pronunciation mode into the electronic device In 20, the electronic device 20 records. The number of words learned and the number of assessment passes can be recorded according to the operation of the user 10 on the electronic device 20. For example, the user 10 evaluates a word. If the pronunciation of the user 10 matches the pronunciation of the word stored in the electronic device 20, the assessment Increase the number of passes by 1. If the user 10 has learned a word once, the learning frequency of the word is increased by one. In some embodiments, the number of assessment passes may be continuously cumulative, that is, if any of the assessments fails, the number of assessment passes will be cleared and recalculated. The number of learning times may not be continuously cumulative. In other embodiments, the number of assessment passes may not be continuously cumulative.

According to the above record of the word by the electronic device 20, the associated historical spelling records of each spelling partition in the database can be updated. In some embodiments, the association history spelling records include the familiarity records of the associated words of each spelling section. The familiarity record includes the number of times each learning type is performed, the pronunciation method, and the last time the related operation is performed, etc. The number of times each learning type is performed, such as the number of learning times, the number of assessment passes, and so on. If a related word of the spelling section has the above familiarity record, the related word is the related historical spelling word of the spelling section. Conversely, if a spelling section has multiple related words, each related word None of the above familiarity records, the spelling section does not have associated historical spelling words. Please refer to Table 1. Table 1 exemplarily shows the familiarity records of the related words in the spelling partition, where “GP” represents the number of passing examinations and “L” represents the number of learning. difficult, difficult and difficult are spelling partitions

Spelling words related to history.

In some of these embodiments, the segmented spelling results containing historical spelling experience may be spelling divisions in various combinations of spelling divisions of the target word, and each associated historical spelling word of each spelling division. In some other embodiments, the electronic device 20 may select a preferred spelling section and its associated historical spelling words from multiple spelling sections and multiple associated historical spelling words according to a preset filtering rule, and according to The preferred spelling partition and its associated historical spelling words prompt the user 10.

Among them, the screening rule may be screening according to the screening rule in a logical selection manner, or it may be combined with a mathematical operation method to quantitatively calculate the screening rule to obtain a screening result.

Among them, the filtering rules can be preset according to the specific application situation. The following examples illustrate the filtering rules. For example, the screening rule may be one or more of a borrowable degree rule, a comprehensive spelling combination correct probability rule, a familiarity rule, and a random rule.

Wherein, the borrowable degree rule is to select a spelling partition combination whose borrowable spelling experience coverage value meets a preset coverage threshold from at least one spelling partition combination, and the borrowable spelling experience coverage value is a spelling partition combination The total number of letters in each spelling section with associated historical spelling words in the combination, or the ratio of the total number of letters in each spelling section with associated historical spelling words in the combination of spelling sections to the number of letters in the target word, or The ratio of the total number of spelling partitions of each spelling partition with associated historical spelling words in the combination of spelling partitions to the number of spelling partitions of the target word. For example, spelling partition combinations

In, only

If the spelling words differential and diffuse with associated history are calculated by the first method above, the borrowed spelling experience coverage value of the spelling partition combination is 3. If calculated by the second method above, the coverage of the borrowable spelling experience coverage of the spelling partition combination is 3/9, which is about 0.33. If calculated by the third method above, the spelling partition combination has a total of 3 partitions,

Only one of the partitions is occupied, so the borrowable spelling experience coverage value of the spelling partition combination is 1/3, which is about 0.33.

The borrowable spelling experience coverage value represents the degree of coverage of the borrowed historical spelling experience that accounts for the entire strange word. The higher the coverage value, the higher the degree of coverage. The borrowable degree rule focuses on the extent to which the borrowable historical spelling experience covers the entire strange word to the highest possible degree. For example, in the spelling partition combination of the strange word different and its associated historical spelling words, the first group is

Only the spelling section

Spelling words differential and diffuse with associated history. And the second group is

In addition to the spelling partition

In addition to the related historical spelling words differ and diffuse, there are also spelling partitions

Spelling word component with associated history. Therefore, the first group has only

The spelling words with associated history and the overall borrowable spelling experience coverage value is 3, while the second group has

with

The coverage value of the borrowable spelling experience that all have related historical spelling words to form the whole group is 6. When using the borrowing degree rules to filter spelling partition combinations, between the first group and the second group, a second group of spelling partition combinations with a higher coverage value of borrowable spelling experience can be selected.

In cases where the focus is on the extent to which the borrowable historical spelling experience covers the entire strange word, the borrowable degree rule can be used first.

The rule for correct probability of the comprehensive spelling combination is to select a spelling partition combination whose correct probability of the comprehensive spelling combination meets a preset threshold from at least one spelling partition combination, and the correct probability of the comprehensive spelling combination is based on the spelling partition combination The structural pronunciation of each spelling partition is obtained with a correct probability. For example, the correct probability of the comprehensive spelling combination can be obtained by calculating the probability of correct pronunciation of the structural pronunciation of each spelling partition in the spelling partition combination using a mathematical average algorithm. Use simple arithmetic average algorithm to calculate spelling partition combination

The correct probability of the comprehensive spelling combination is shown in the example. The correct probability of the comprehensive spelling combination is (100% + 99.81% + 100%) / 3 = 0.9993.

The spelling partition combination with a high probability of comprehensive spelling combination can essentially improve the probability of correct pronunciation of the target word from the spelling structure. The rule of correct probability of combined spelling combination focuses on improving the probability of correct pronunciation of the target word essentially from the structure.

The familiarity rule is to select, from at least one spelling partition combination and its associated historical spelling words, at least one spelling partition combination and its associated historical spelling with a relatively high degree of comprehensive familiarity in each spelling partition For reading words, the comprehensive familiarity degree is obtained based on the familiarity of the historical spelling words associated with each spelling partition. For example, the comprehensive familiarity degree can be obtained by adding the familiarity degree of historical spelling words associated with each spelling section. In some embodiments, the familiarity of the related historical spelling words may be determined solely by the familiarity type or the number of times the type is performed. In other embodiments, the familiarity type and the number of times of the type may also be determined at the same time.

The familiarity type and the number of types of times the associated historical spelling words can be obtained based on the familiarity record of the associated historical spelling words. In some of these embodiments, the familiarity types include full reading, spelling, and learning. The pronunciation method is full reading and the number of assessment passes is non-zero. The full pronunciation class is pronunciation and the reading method is non-zero. The class is spelling, and the number of passing exams is 0, but the number of learning is not 0 is the class.

Among them, the familiarity type, and the relationship between the familiarity degree and the familiarity type and the number of times the type is performed can be defined according to different application scenarios. In some of these embodiments, the familiarity of the whole reading class is greater than the familiarity of the spelling class, and the familiarity of the spelling class is greater than the familiarity of the learning class. The more the type is performed, the higher the familiarity. In some applications, if the familiarity of the assessment is greater than the familiarity of learning, you can set the relationship between the number of assessments passed and the number of learning. For example, five learning times can be equivalent to The number of passes for an assessment.

When the familiarity of the associated historical spelling words depends solely on the number of times the associated historical spelling words are used, when using the familiarity rules to select the spelling partition combination and its associated historical spelling words, you can use the number of times of each type (for example The number of assessment passes and the number of learnings) and the corresponding relationship between them are given priority to select the associated historical spelling word type of each spelling partition to perform more frequent spelling partition combinations and their associated historical spelling words.

When the familiarity degree depends solely on the familiarity type, when using the familiarity rules to select the spelling partition combination and its associated historical spelling words, you can prefer to select the associated historical spelling words of each spelling partition that contain more whole reading classes Associated historical spelling words and their combination of spelling partitions, followed by selecting the associated historical spelling words of each spelling partition containing more historical spelling related words and their spelling partition combinations, and finally selecting each spelling partition The related historical spelling words include more related historical spelling words in the learning category and their combination of spelling partitions.

In cases where the familiarity degree depends on both the familiarity type and the number of times the type is performed, you can set the preference order of the familiarity type and the number of times of the type, for example, the priority of the familiarity type is greater than the priority of the type of number of times. Then, when using the familiarity rules to select spelling partition combinations and their associated historical spelling words, you can prefer to select the associated historical spelling words of each spelling partition that contain more full-read related history spelling words and their spelling partition combinations , If the associated historical spelling words of each spelling partition contain the same number of whole reading classes, you can choose the combination of spelling partitions and their associated historical spelling words that are performed more frequently according to the number of types. Repeat.

The above familiarity rule is a logical screening method, which needs to preset the preference level of the familiarity type or the number of times of type, but in some applications, the familiarity type and the number of times of type have the same priority level. Considering the familiarity type and the number of times the type is performed, a numerical quantization method can be used to indicate the familiarity level of related historical spelling words with the familiarity index. When using the familiarity rules to select the spelling partition combination and its associated historical spelling words, you can select the spelling partition combination and its associated historical spelling words with a higher sum of familiarity index. The familiarity index can be obtained by calculating the number of times according to the familiarity type and type. You can define the familiarity type coefficient and the value of the number of times coefficient in advance, and then calculate the familiarity index based on the familiarity type coefficient and type. For example, the familiarity index is the product of the familiarity type coefficient and the type frequency coefficient.

Among them, in some applications, as the number of studies or assessments increases, the "benefit" of familiarity also increases. However, as the number of types progressed to a certain number of times, the enhancement rate of the "benefit" presented therein will become less and less. In order to reflect the enhanced “benefit” of the familiarity degree by the number of times of type progress, the coefficient of the number of times of type progress can be determined according to the non-linear exchange table set in advance. Table 4 exemplarily shows the relationship between the value of the number of times the type progresses and the number of times the type progresses. Among them, the coefficient of the number of times of each type (for example, the coefficient of the number of times of passing the assessment and the coefficient of the number of times of learning, etc.) may use the same non-linear swap table, or may define different non-linear swap tables separately. In the case where the number of assessments is more important than the number of learnings, different nonlinear swap tables are defined respectively. When each nonlinear swap table has the same type frequency, the value of the coefficient of the number of assessment passes is greater than the value of the coefficient of the number of learning times.

Table 4

Types performed	Type Progression Coefficient
1	0.63161
2	0.86429
3	0.95
4	0.98158
5	0.99321

6	0.9975
7	0.99908
8	0.99966
9	0.99988
10	0.99995
11	0.99998
12	0.99999
13	1
14	1

In some other embodiments, the type progression coefficient can also be obtained by a nonlinear function, where the nonlinear function is, for example, a polyline function, a logarithmic function, an exponential function, a power function, a polynomial function, and so on. The coefficient of the number of times of each type can be expressed by the same function or by different functions. Taking the number of times performed by each type as an example of the same exponential function, the coefficient of the number of times performed by each type can be expressed as 1-exp (-Ti / A) or 1-exp (-Ti * B), where exp () is returned by the exponential function e to the power, i is the i-th item, Ti is the number of times the i-th spelling partition is performed, A and B are preset limit constants, which can be set in advance to determine which value T _i reaches, It has caused the corresponding type to perform the number of times coefficient value to reach a preset near-top value situation.

The familiarity type coefficient Wi (for example, the full reading coefficient WiF, the spelling coefficient WiP, and the learning coefficient WiL) can be set in advance. The familiarity type coefficient Wi can be set according to the contribution of the familiarity type to the familiarity level in different applications. Taking the familiarity type coefficient Wi including the full reading coefficient WiF, the spelling coefficient WiP and the learning coefficient WiL as an example, there is little difference in familiarity from the spelling to the full reading state, while the learning category is relative to the spelling Class and full-read class, the degree of familiarity is quite different. Therefore, you can set WiP to be less than Wi, and WiL to be much smaller than WiP and WiF, so that when the familiarity index is calculated by the familiarity type coefficient and the number of times coefficient, it will not affect the whole reading class and spelling because of the large number of learnings. The importance of reading classes. For example, WiF, WiP, and WiL can be set to 1, 0.64346, and 0.03163, respectively.

The familiarity rule is to provide users with more familiar associated historical spelling words to further prevent users 10 from borrowing historical spelling memory by mistake. In addition, it also helps the user 10 to have higher spelling proficiency, pronunciation familiarity and longer memory persistence for the related phonetic regions of strange words.

The above rules can be selected or applied according to specific application conditions, and the preset coverage threshold and the preset threshold can be set according to specific application conditions. It should be noted that, because the rule of correct probability of comprehensive spelling combination is applicable to the selection of combination of spelling partitions, it is not applicable to the selection of related historical spelling words. Therefore, in the occasions where the screening requirements of related historical spelling words are high, you can choose the familiarity rules, or combine the comprehensive probability spelling correct probability rules, the borrowability degree rules, and the familiarity rules. When the target word has only one spelling partition combination, it is not necessary to filter the spelling partition combination. In this case, you can use the familiarity rules alone to filter the related historical spelling words.

In other embodiments of the present invention, in order to comprehensively consider factors such as the value of the spelling experience coverage that can be borrowed, the probability of correct pronunciation of the structural pronunciation of each spelling section, and the familiarity of the historical spelling words associated with each spelling section, a numerical value can be used A quantitative method is used to obtain the comprehensive index of spelling partition experience of each spelling partition combination and its associated historical spelling words, and the preferred level of spelling partition combination and its associated historical spelling words is characterized by the comprehensive index of spelling partition experience. Then, from the at least one spelling partition combination and its associated historical spelling words, at least one spelling partition combination and its associated historical spelling word with a high spelling partition experience comprehensive index are selected. The comprehensive index of spelling division experience can be obtained based on the borrowable spelling experience coverage value of each spelling division, the probability of correct pronunciation of each spelling division structure, and the familiarity of the historical spelling words associated with each spelling division.

For example, in a situation where the degree of familiarity depends on the type of familiarity, the comprehensive index F of the spelling division experience is:

Where n is the total number of spelling partitions of the target word, S _i is the number of letter strings in the spelling partition with associated historical spelling words, P _i is the probability of correct pronunciation of the structural pronunciation of the i-th spelling partition, W _i represents the familiarity type coefficient of the historical spelling words associated with the i-th spelling partition, and L represents the total number of letter strings of the target word, so S _i / L is the borrowable spelling coverage value of the i-th spelling partition. In this embodiment, the spelling experience coverage value can be borrowed to calculate the ratio of the sum of the number of letters in each spelling section with associated historical spelling words in the spelling section combination to the number of letters in the target word. In the embodiment, the borrowable spelling experience coverage value of the i-th spelling partition can also be the sum of the number of letters in each spelling partition with associated historical spelling words in the spelling partition combination, or the association in the spelling partition combination The ratio of the total number of spelling partitions of each spelling partition of the historical spelling word to the number of spelling partitions of the target word is calculated.

Where familiarity depends on both the familiarity type and the number of times the type is performed, the comprehensive index F of the spelling division experience is:

Where, T _i represents the number of times the type of spelling words associated with the spelling partition of the ith item is performed, and A is a preset limit constant;

or,

Among them, if in some applications, the preferred possibility spelling combination and the related historical spelling word combination are considered in a non-intimately dependent factor relationship, each factor can be added, multiplied or The relationship is a combination of multiplication and multiplication. When it is necessary to consider the value of borrowing spelling experience coverage, the probability of correct pronunciation of the structural pronunciation of each spelling section, and the familiarity of the historical spelling words associated with each spelling section, you can add the factors in the above formula according to the degree of importance Set weights. E.g

or

Among them, a, b, c are preset weight values.

Among the factors that can be borrowed spelling experience coverage value, familiarity, and structural pronunciation correct probability, each factor has its own advantages and disadvantages. The benefit of the spelling division experience comprehensive index F is that it can balance the advantages and disadvantages of various factors at the same time, and combine the spelling division combination and its associated historical spelling words to form each district associative spelling containing historical spelling experience Program combinations, and choose the best combination.

In the embodiment, the combination of the region-associative spelling scheme may be expressed in the manner of spelling block sequence. For example, a spelling partition combination "dif-fer-ent", each spelling partition selects a related historical spelling word to form a spelling block unit, and the spelling block units are arranged in order to form a partition association spelling Reading program combination. For example, one partition association spelling portfolio expression may be "dif (dif ficult) -fer ( re fer ence) -ent (compon ent)".

For specific applications, for each spelling partition combination and each associated historical spelling word of each spelling partition in the spelling partition combination, the combination of each district ’s associative spelling scheme can be split and the corresponding synthesis of the spelling partition experience can be calculated Index F. Then, choose the combination of the associative spelling schemes with the highest comprehensive index F in the spelling division experience. The following uses one of the spelling partition combinations “dif-feren-t” in Table 1 as an example to explain how to calculate the comprehensive index F of the spelling partition experience corresponding to each partition associative spelling scheme combination of each spelling partition combination.

The spelling partition combination "dif-feren-t" in Table 1 shows the related historical spelling words of each spelling partition, where "dif" has 3 related historical spelling words differential, differential and differential, " "Feren" has 2 related historical spelling words reference and preference. Therefore, from this single spelling partition combination, a total of 6 partition association spelling plan combinations and their corresponding spelling partition experience comprehensive index F are generated, and they all belong to the same possible spelling partition combination "dif-feren-t" (Please refer to Table 5). The combination of the six district Lenovo spelling schemes are as follows:

"Dif (difficult) -feren (reference) -t", "dif (difficult) -feren (preference) -t",

"Dif (difficulty) -feren (reference) -t", "dif (difficulty) -feren (preference) -t",

"Dif (diffident) -feren (reference) -t", "dif (diffident) -feren (preference) -t".

Table 5 shows a calculation example of the comprehensive index F of the spelling partition experience corresponding to the combination of the six partition associative spelling schemes, which is calculated and displayed according to the formula F2. Among them, the comprehensive index of spelling division experience corresponding to "dif (difficult) -feren (preference) -t" F = 3 * 100% * (1-exp (-1 / 1.0014)) = 0.435891 is the highest. Similarly, for the other spelling partition combinations and their associated historical spelling words in Table 1, the associated partition associative spelling scheme combinations are also generated according to the above method, and the corresponding comprehensive index F of the spelling partition experience is calculated. Assuming that the other possible spelling partition combinations and their associated historical spelling words, the corresponding combined index F of each spelling partition experience index F is lower than 0.435891, then "dif (difficult) -feren (preference) -t "can be used as the best combination of associative spelling scheme. The electronic device 20 may prompt the user 10 according to the best combination of associative spelling schemes in the region.

table 5

The embodiment of the present invention obtains the target word by acquiring at least one spelling partition combination of the target word and based on the at least one spelling partition combination and the related historical spelling records of the spelling partition in the spelling partition combination Partition spelling results including historical spelling experience. When learners are learning strange words, they can provide section spelling tips including correct historical spelling experience to avoid learners from borrowing historical spelling memory by mistake.

Correspondingly, an embodiment of the present invention also provides a word learning device that borrows historical spelling experience. The word learning device that borrows historical spelling experience can be used in the electronic device 20 in FIG. 1, as shown in FIG. 3, The word learning device 300 borrowing historical spelling experience includes:

The spelling partition combination acquisition module 301 is configured to acquire at least one spelling partition combination of the target word, and the spelling partition combination includes at least one spelling partition;

The spelling result acquisition module 302 is configured to obtain the regional spelling of the target word including the historical spelling experience based on the at least one spelling partition combination and the associated historical spelling records of the spelling partitions in the spelling partition combination Read the results.

In some embodiments of the word learning device 300 borrowing historical spelling experience, the related history spelling record includes at least one related history spelling word and a familiarity record of the at least one related history spelling word.

Referring to FIG. 4, the spelling result acquisition module 302 includes:

The filtering module 3021 is configured to select from the at least one spelling partition combination based on the at least one spelling partition combination and the associated historical spelling records of the spelling partition in the spelling partition combination according to a first preset filtering rule Filter out the regional spelling results that contain the related historical spelling words.

Wherein, in some embodiments of the word learning device 300 borrowing historical spelling experience, the first preset screening rule includes at least one of the following rules:

You can borrow the rules of the degree, the rules of the correct probability of combining spelling combinations, the rules of the degree of familiarity, and the random rules;

Wherein, the borrowable degree rule is to select, from at least one spelling partition combination, a spelling partition combination whose borrowable spelling experience coverage value meets a preset coverage threshold, and the borrowable spelling experience coverage value is spelling The sum of the number of letters in each spelling partition with associated historical spelling words in the partition combination, or the ratio of the sum of the letters in each spelling partition with associated historical spelling words in the spelling partition combination to the number of letters in the target word Or the ratio of the total number of spelling partitions of each spelling partition with associated historical spelling words in the spelling partition combination to the number of spelling partitions of the target word;

The rule for correct probability of the comprehensive spelling combination is to select, from at least one spelling partition combination, a spelling partition combination whose correct probability of meeting the preset threshold value is based on the spelling partition The probability of the structural pronunciation of each spelling partition in the combination is obtained correctly;

The familiarity rule is to select, from at least one spelling partition combination and its associated historical spelling words, at least one spelling partition combination and its associated historical spelling with a relatively high degree of comprehensive familiarity in each spelling partition Read the word, the comprehensive familiarity degree is obtained based on the familiarity degree of the associated historical spelling words in each spelling section, and the familiarity degree of the associated historical spelling words depends on the familiarity type and / or type of the associated historical spelling words The number of times of execution, the type of familiarity and the number of times of type determination are determined according to the familiarity record of the associated historical spelling words;

Random rule: From at least one spelling partition combination, randomly select a spelling partition combination.

In some embodiments of the word learning device 300 borrowing historical spelling experience, the first preset filtering rule is to select spelling from the at least one spelling partition combination and its associated historical spelling words At least one spelling partition combination with a high partition experience comprehensive index and its associated historical spelling words, the spelling partition experience comprehensive index is obtained based on the spelling experience index of each spelling partition, and the spelling of each spelling partition The experience index is obtained based on the borrowable spelling experience coverage value of each spelling section, the probability of correct pronunciation of the structural pronunciation of each spelling section, and the familiarity of the historical spelling words associated with each spelling section. The degree depends on the familiarity type and / or type progression times of the associated historical spelling words, and the familiarity type and type progression frequency are determined according to the familiarity record of the associated historical spelling words.

Among them, in some embodiments of the word learning device 300 borrowing historical spelling experience, the familiarity of the related historical spelling words depends on the familiarity type and the number of times the type of related historical spelling words are performed. The degree type includes at least one of the whole reading class, the spelling class and the learning class;

The familiarity index of the related historical spelling words is the product of the type coefficient of the familiarity of the related historical spelling words and the coefficient of the number of times the type is performed, wherein the familiarity index of the related historical spelling words reflects the related historical spelling words Familiarity level, the greater the familiarity index, the higher the familiarity level;

Among them, the familiarity type coefficient of the whole reading class is greater than the familiarity type coefficient of the spelling class, and the familiarity type coefficient of the spelling class is greater than the familiarity type coefficient of the learning class;

The type of the number of times of progress coefficient increases as the number of times of progress increases, and the speed of increase becomes slower as the number of times of progress increases.

Among them, in some embodiments of the word learning device 300 borrowing historical spelling experience, the spelling partition combination acquisition module 301 is specifically used to:

Obtain at least one possible spelling partition combination of the target word;

or,

Obtain the best spelling partition combination of the target word, merge and / or subdivide each spelling partition in the best spelling partition combination to obtain at least one spelling partition combination of the target word.

It should be noted that the above word learning device for borrowing historical spelling experience can execute the word learning method for borrowing historical spelling experience provided by the embodiment of the present invention, and has corresponding function modules and a word learning method for executing the borrowing history spelling experience. Beneficial effect. For technical details that are not described in detail in the device embodiments, please refer to the word learning method for borrowing historical spelling experience provided by the embodiments of the present invention.

As shown in FIG. 5, an embodiment of the present invention also provides an electronic device 20. The electronic device 20 includes:

One or more processors 21 and a memory 22, one processor 21 is taken as an example in FIG. 5. The processor 21 and the memory 22 may be connected through a bus or in other ways. In FIG. 5, a bus connection is used as an example.

The memory 22 is a non-volatile computer-readable storage medium, and can be used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as borrowing historical spelling experience words in the embodiments of the present invention. Program instructions / units corresponding to the learning method (for example, the spelling partition combination acquisition module 301 and the spelling result acquisition module 302 shown in FIG. 3). The processor 21 executes various functional applications and data processing of the electronic device by running the non-volatile software programs, instructions, and units stored in the memory 22, that is, word learning that borrows historical spelling experience from the above method embodiments is implemented method.

The memory 22 may include a storage program area and a storage data area, where the storage program area may store an operating system and application programs required for at least one function; the storage data area may store data created according to the use of electronic devices, and the like. In addition, the memory 22 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 22 may optionally include memories remotely provided with respect to the processor 21, and these remote memories may be connected to the electronic device through a network. Examples of the above network include but are not limited to the Internet, intranet, local area network, mobile communication network, and combinations thereof.

The one or more units are stored in the memory 22, and when executed by the one or more processors 21, execute the word learning method of borrowing historical spelling experience in any of the above method embodiments, for example, executing The method steps 101-102 in FIG. 2 described above implement the functions of the modules 301-302 shown in FIG. 3 and the modules 301, 302, and 3021 shown in FIG.

The above electronic device can execute the word learning method borrowed from the historical spelling experience provided by the embodiments of the present invention, and has the corresponding function modules and beneficial effects of the execution method. For technical details that are not described in detail in the embodiment of the electronic device, refer to the method provided in the embodiment of the present invention.

An embodiment of the present invention provides a non-volatile computer-readable storage medium that stores computer-executable instructions that are executed by one or more processors, such as in FIG. 5 A processor 21 may enable the one or more processors to execute the word learning method of borrowing historical spelling experience in any of the above method embodiments, for example, performing the method steps 101-102 of FIG. 2 described above, The functions of the modules 301-302 shown in FIG. 3 and the modules 301, 302, and 3021 shown in FIG. 4 are implemented.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, not to limit them; under the idea of the present application, the technical features in the above embodiments or different embodiments may also be combined, The steps can be implemented in any order, and there are many other variations of the different aspects of the present application as described above. For simplicity, they are not provided in the details; although the present application has been described in detail with reference to the foregoing embodiments, the ordinary The skilled person should understand that they can still modify the technical solutions described in the foregoing embodiments, or equivalently replace some of the technical features; and these modifications or replacements do not deviate from the essence of the corresponding technical solutions in the implementation of this application. Examples of technical solutions.

Claims (10)

1. A word learning method borrowing historical spelling experience for electronic devices, characterized in that the method includes:

   Obtaining at least one spelling partition combination of the target word, where the spelling partition combination includes at least one spelling partition;

   Based on the at least one spelling partition combination and the associated historical spelling records of the spelling partitions in the spelling partition combination, a partition spelling result containing historical spelling experience of the target word is obtained.
2. The method according to claim 1, wherein the related history spelling records include at least one related history spelling word and the familiarity record of the at least one related history spelling word;

   Based on the at least one spelling partition combination and the associated historical spelling records of the spelling partitions in the spelling partition combination, obtaining the partition spelling result of the target word that includes the historical spelling experience includes:

   Based on the at least one spelling partition combination and the associated history spelling records of the spelling partitions in the spelling partition combination, according to a first preset filtering rule, filtering out the association history from the at least one spelling partition combination Partition spelling results of spelling words.
3. The method according to claim 2, wherein the first preset screening rule includes at least one of the following rules:

   You can borrow the rules of the degree, the rules of the correct probability of combining spelling combinations, the rules of the degree of familiarity, and the random rules;

   Wherein, the borrowable degree rule is to select, from at least one spelling partition combination, a spelling partition combination whose borrowable spelling experience coverage value meets a preset coverage threshold, and the borrowable spelling experience coverage value is spelling The sum of the number of letters in each spelling partition with associated historical spelling words in the partition combination, or the ratio of the sum of the letters in each spelling partition with associated historical spelling words in the spelling partition combination to the number of letters in the target word Or the ratio of the total number of spelling partitions of each spelling partition with associated historical spelling words in the spelling partition combination to the number of spelling partitions of the target word;

   The rule for correct probability of the comprehensive spelling combination is to select, from at least one spelling partition combination, a spelling partition combination whose correct probability of meeting the preset threshold value is based on the spelling partition The probability of the structural pronunciation of each spelling partition in the combination is obtained correctly;

   The familiarity rule is to select, from at least one spelling partition combination and its associated historical spelling words, at least one spelling partition combination and its associated historical spelling with a relatively high degree of comprehensive familiarity in each spelling partition Read the word, the comprehensive familiarity degree is obtained based on the familiarity degree of the associated historical spelling words in each spelling section, and the familiarity degree of the associated historical spelling words depends on the familiarity type and / or type of the associated historical spelling words The number of times of execution, the type of familiarity and the number of times of type determination are determined according to the familiarity record of the associated historical spelling words;

   Random rule: From at least one spelling partition combination, randomly select a spelling partition combination.
4. The method according to claim 2, wherein the first preset screening rule is to select a higher comprehensive index of spelling division experience from the at least one spelling division combination and its associated historical spelling words At least one combination of spelling partitions and their associated historical spelling words, the comprehensive index of spelling partition experience is obtained based on the spelling experience index of each spelling partition, and the spelling experience index of each spelling partition is based on each spelling The borrowed spelling experience coverage value of the partition, the probability of correct pronunciation of the structural pronunciation of each spelling partition, and the familiarity of the historical spelling words associated with each spelling partition are obtained. The familiarity type and / or type progression times of the historical spelling words are determined according to the familiarity record of the associated historical spelling words.
5. The method according to claim 3 or 4, characterized in that the familiarity of the associated historical spelling words depends on the familiarity type of the associated historical spelling words and the number of types performed, and the familiarity type includes integral At least one of reading, spelling and learning;

   The familiarity index of the related historical spelling words is the product of the type coefficient of the familiarity of the related historical spelling words and the coefficient of the number of times the type is performed, wherein the familiarity index of the related historical spelling words reflects the related historical spelling words Familiarity level, the greater the familiarity index, the higher the familiarity level;

   Among them, the familiarity type coefficient of the whole reading class is greater than the familiarity type coefficient of the spelling class, and the familiarity type coefficient of the spelling class is greater than the familiarity type coefficient of the learning class;

   The type of the number of times of progress coefficient increases as the number of times of progress increases, and the speed of increase becomes slower as the number of times of progress increases.
6. The method according to claim 1, wherein the at least one spelling partition combination for acquiring the target word comprises:

   Obtain at least one possible spelling partition combination of the target word;

   or,

   Obtain the best spelling partition combination of the target word, merge and / or subdivide each spelling partition in the best spelling partition combination to obtain at least one spelling partition combination of the target word.
7. A word learning device borrowing historical spelling experience for electronic equipment, characterized in that the word learning device includes:

   A spelling partition combination acquisition module, configured to acquire at least one spelling partition combination of the target word, and the spelling partition combination includes at least one spelling partition;

   The spelling result acquisition module is used to obtain the segmentation spelling of the target word containing the historical spelling experience based on the at least one spelling division combination and the associated historical spelling records of the spelling division in the spelling division combination result.
8. The apparatus according to claim 7, wherein the related history spelling records include at least one related history spelling word and the familiarity record of the at least one related history spelling word;

   The spelling result acquisition module includes:

   The filtering module is configured to select from the at least one spelling partition combination based on the at least one spelling partition combination and the associated historical spelling records of the spelling partition in the spelling partition combination according to a first preset filtering rule Filter out the segmentation spelling results that contain related historical spelling words.
9. An electronic device, characterized in that the electronic device includes:

   At least one processor; and,

   A memory communicatively connected to the at least one processor; wherein,

   The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute any one of claims 1-6 method.
10. A non-volatile computer-readable storage medium, characterized in that the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by an electronic device, the electronic device is allowed to execute rights The method according to any one of claims 1 to 6.

Images