A data processing method and device for constructing a domain dictionary

Abstract

The present invention discloses a data processing method and device for constructing a domain dictionary, the method comprising: obtaining information of material to be processed; the material information to be processed includes marketing text information, and / or song information to be processed, and / or rhyme word table information; segmenting and filtering the material information to be processed to obtain a dictionary to be used; the dictionary to be used is used to screen words in the material text; segmenting and filtering the dictionary to be used to obtain domain keyword information; the keyword information is strongly related to the advertising field to which the material information to be processed belongs; the domain keywords are used to construct the domain dictionary. It can be seen that the present invention can obtain domain keyword information for constructing a domain dictionary through comprehensive processing such as segmentation and filtering processing and segmentation and screening processing of the material information to be processed, which is conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thus improving the generation efficiency of advertising lyrics and reducing production costs.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a data processing method and device for constructing a domain dictionary. Background Art

[0002] Converting popular pop songs into advertising lyrics is an important way for current advertising promotion to adapt to the development of new media such as elevator advertising. This process requires the use of intelligent models to learn the language style, social trends and emotional expressions behind the promotional songs to generate catchy advertising lyrics. However, in the current song generation process, there are generally problems such as high requirements for song materials and high production costs of advertising lyrics. Therefore, it is particularly important to provide a data processing method and device for constructing a domain dictionary to reduce the requirements for song materials, improve the processing efficiency of song materials, and thus improve the generation efficiency of advertising lyrics and reduce production costs. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a data processing method and device for constructing a domain dictionary, which can obtain domain keyword information for constructing a domain dictionary through comprehensive processing such as segmentation and filtering processing and segmentation and screening processing of the material information to be processed, which is conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and then improving the generation efficiency of advertising lyrics and reducing production costs.

[0004] In order to solve the above technical problems, a first aspect of an embodiment of the present invention discloses a data processing method for constructing a domain dictionary, the method comprising:

[0005] Acquire the material information to be processed; the material information to be processed includes marketing text information, and / or song information to be processed, and / or rhyme word list information;

[0006] Segment and filter the material information to be processed to obtain a dictionary to be used; the dictionary to be used is used to filter words in the material text;

[0007] The dictionary to be used is segmented and screened to obtain domain keyword information; the keyword information is strongly related to the advertising domain to which the material information to be processed belongs; the domain keywords are used to construct a domain dictionary.

[0008] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the segmentation and filtering of the to-be-processed material information to obtain a dictionary to be used includes:

[0009] Performing identification processing on the song information to be processed to obtain initial lyrics text information;

[0010] The initial lyrics text information is segmented and filtered to obtain a dictionary to be used.

[0011] As an optional implementation, in the first aspect of the embodiment of the present invention, the segmentation and filtering of the initial lyrics text information to obtain a dictionary to be used includes:

[0012] Performing character segmentation and line separation processing on the initial lyrics text information to obtain first intermediate lyrics text information;

[0013] Performing calculation processing on the first intermediate lyrics text information to obtain second intermediate lyrics text information;

[0014] The second intermediate lyrics text information is screened and processed using a preset stop word list to obtain third intermediate lyrics text information; the third intermediate lyrics text information includes a plurality of lyrics text segments;

[0015] The third intermediate lyrics text information is processed by calculation, screening and sorting to obtain a dictionary to be used.

[0016] As an optional implementation, in the first aspect of the embodiment of the present invention, the third intermediate lyrics text information is subjected to calculation, screening and sorting processing to obtain a dictionary to be used, including:

[0017] Performing information entropy calculation processing on the third intermediate lyrics text information to obtain first text entropy information;

[0018] Using a preset entropy threshold to filter the first text entropy information to obtain second text entropy information;

[0019] Sorting the second text entropy information to obtain text sequence information; the sorting is performed based on frequency information corresponding to the lyrics text segments;

[0020] The text sequence information is processed to obtain a dictionary to be used.

[0021] As an optional implementation, in the first aspect of the embodiment of the present invention, the segmentation and screening of the dictionary to be used to obtain the domain keyword information includes:

[0022] Using a preset word vector model to process the dictionary to be used and the marketing text information to obtain word vector information;

[0023] Using a preset first word segmentation tool to process the dictionary to be used and the marketing text information to obtain field word information to be used;

[0024] The word vector information and the to-be-used domain word information are converted and screened to obtain domain keyword information.

[0025] As an optional implementation, in the first aspect of the embodiment of the present invention, the conversion and screening of the word vector information and the to-be-used domain word information to obtain domain keyword information includes:

[0026] Using the word vector information, the to-be-used domain word information is converted to obtain domain word vector information;

[0027] Performing clustering calculation on the domain word vector information to obtain word distance information;

[0028] The word distance information is grouped and screened to obtain domain keyword information.

[0029] As an optional implementation, in the first aspect of the embodiment of the present invention, after the segmentation and screening process is performed on the dictionary to be used to obtain the domain keyword information, the method further includes:

[0030] Processing the rhyming word list information to obtain a homophonic word list;

[0031] The preset second word segmentation tool is used to process the homophonic list, the field keyword information and the song information to be processed to obtain training text information; the training text information is used to train the format restriction model; the format restriction model is used to constrain the format of the advertisement lyrics.

[0032] A second aspect of an embodiment of the present invention discloses a data processing device for constructing a domain dictionary, the device comprising:

[0033] An acquisition module, used for acquiring material information to be processed; the material information to be processed includes marketing text information, and / or song information to be processed, and / or rhyme word table information;

[0034] The first processing module is used to segment and filter the material information to be processed to obtain a dictionary to be used; the dictionary to be used is used to filter words in the material text;

[0035] The second processing module is used to segment and filter the dictionary to be used to obtain domain keyword information; the keyword information is strongly related to the advertising domain to which the material information to be processed belongs; the domain keywords are used to construct a domain dictionary.

[0036] As an optional implementation, in the second aspect of the embodiment of the present invention, the first processing module includes a first processing submodule and a second processing submodule, wherein:

[0037] The first processing submodule is used to identify and process the song information to be processed to obtain initial lyrics text information;

[0038] The second processing submodule is used to segment and filter the initial lyrics text information to obtain a dictionary to be used.

[0039] As an optional implementation, in the second aspect of the embodiment of the present invention, the second processing submodule performs segmentation and filtering on the initial lyrics text information to obtain a dictionary to be used, including:

[0040] Performing character segmentation and line separation processing on the initial lyrics text information to obtain first intermediate lyrics text information;

[0041] Performing calculation processing on the first intermediate lyrics text information to obtain second intermediate lyrics text information;

[0042] The second intermediate lyrics text information is screened and processed using a preset stop word list to obtain third intermediate lyrics text information; the third intermediate lyrics text information includes a plurality of lyrics text segments;

[0043] The third intermediate lyrics text information is processed by calculation, screening and sorting to obtain a dictionary to be used.

[0044] As an optional implementation, in the second aspect of the embodiment of the present invention, the second processing submodule performs calculation, screening and sorting processing on the third intermediate lyrics text information to obtain a dictionary to be used, including:

[0045] Performing information entropy calculation processing on the third intermediate lyrics text information to obtain first text entropy information;

[0046] Using a preset entropy threshold to filter the first text entropy information to obtain second text entropy information;

[0047] Sorting the second text entropy information to obtain text sequence information; the sorting is performed based on frequency information corresponding to the lyrics text segments;

[0048] The text sequence information is processed to obtain a dictionary to be used.

[0049] As an optional implementation, in the second aspect of the embodiment of the present invention, the second processing module performs segmentation and screening processing on the dictionary to be used, and obtains the field keyword information in a specific manner as follows:

[0050] Using a preset word vector model to process the dictionary to be used and the marketing text information to obtain word vector information;

[0051] Using a preset first word segmentation tool to process the dictionary to be used and the marketing text information to obtain field word information to be used;

[0052] The word vector information and the to-be-used domain word information are converted and screened to obtain domain keyword information.

[0053] As an optional implementation, in the second aspect of the embodiment of the present invention, the second processing module performs conversion and screening processing on the word vector information and the to-be-used domain word information to obtain the domain keyword information in a specific manner as follows:

[0054] Using the word vector information, the to-be-used domain word information is converted to obtain domain word vector information;

[0055] Performing clustering calculation on the domain word vector information to obtain word distance information;

[0056] The word distance information is grouped and screened to obtain domain keyword information.

[0057] As an optional implementation, in the second aspect of the embodiment of the present invention, after the second processing module performs segmentation and screening processing on the standby dictionary to obtain field keyword information, the device further includes:

[0058] A third processing module is used to process the rhyme word table information to obtain a homophonic word list;

[0059] The fourth processing module is used to use the preset second word segmentation tool to process the rhyme list, the field keyword information and the song information to be processed to obtain training text information; the training text information is used to train the format restriction model; the format restriction model is used to constrain the format of the advertising lyrics.

[0060] The third aspect of the present invention discloses another data processing device for constructing a domain dictionary, the device comprising:

[0061] A memory storing executable program code;

[0062] a processor coupled to the memory;

[0063] The processor calls the executable program code stored in the memory to execute part or all of the steps in the data processing method for constructing a domain dictionary disclosed in the first aspect of the embodiment of the present invention.

[0064] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the data processing method for constructing a domain dictionary disclosed in the first aspect of the embodiment of the present invention.

[0065] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0066] In an embodiment of the present invention, the material information to be processed is obtained; the material information to be processed includes marketing text information, and / or, song information to be processed, and / or, rhyme word table information; the material information to be processed is segmented and filtered to obtain a dictionary to be used; the dictionary to be used is used to filter words in the material text; the dictionary to be used is segmented and filtered to obtain domain keyword information; the keyword information is strongly related to the advertising field to which the material information to be processed belongs; the domain keywords are used to construct a domain dictionary. It can be seen that the present invention can obtain domain keyword information for constructing a domain dictionary through comprehensive processing such as segmentation and filtering processing and segmentation and filtering processing of the material information to be processed, which is conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thus improving the generation efficiency of advertising lyrics and reducing production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0068] Figure 1 It is a flowchart of a data processing method for constructing a domain dictionary disclosed in an embodiment of the present invention;

[0069] Figure 2 It is a flowchart of another data processing method for constructing a domain dictionary disclosed in an embodiment of the present invention;

[0070] Figure 3 It is a structural schematic diagram of a data processing device for constructing a domain dictionary disclosed in an embodiment of the present invention;

[0071] Figure 4 It is a structural schematic diagram of another data processing device for constructing a domain dictionary disclosed in an embodiment of the present invention;

[0072] Figure 5 A schematic diagram of the structure of another data processing device for constructing a domain dictionary disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0075] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0076] The present invention discloses a data processing method and device for constructing a domain dictionary, which can obtain domain keyword information for constructing a domain dictionary through comprehensive processing such as segmentation and filtering processing and segmentation and screening processing of the material information to be processed, which is conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thus improving the generation efficiency of advertising lyrics and reducing production costs. The following are detailed descriptions.

[0077] Embodiment 1

[0078] See also Figure 1 , Figure 1 1 is a flow chart of a data processing method for constructing a domain dictionary disclosed in an embodiment of the present invention. Figure 1 The data processing method for constructing a domain dictionary described is applied to a data processing system, such as a local server or a cloud server for data processing management for constructing a domain dictionary, and the embodiments of the present invention do not limit this. Figure 1 As shown, the data processing method for constructing a domain dictionary may include the following operations:

[0079] 101. Obtain information of materials to be processed.

[0080] In the embodiment of the present invention, the above-mentioned material information to be processed includes marketing text information, and / or song information to be processed, and / or rhyme word table information, which is not limited in the embodiment of the present invention.

[0081] 102. Segment and filter the material information to be processed to obtain a dictionary to be used.

[0082] In the embodiment of the present invention, the above-mentioned dictionary to be used is used to filter words in the material text.

[0083] 103. Segment and filter the dictionary to be used to obtain domain keyword information.

[0084] In the embodiment of the present invention, the keyword information is strongly related to the advertising field to which the material information to be processed belongs.

[0085] In the embodiment of the present invention, the above-mentioned domain keywords are used to construct a domain dictionary.

[0086] It can be seen that the data processing method for constructing a domain dictionary described in the embodiment of the present invention can obtain domain keyword information for constructing a domain dictionary through comprehensive processing such as segmentation and filtering processing and segmentation and screening processing of the material information to be processed, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and then improve the generation efficiency of advertising lyrics and reduce production costs.

[0087] In an optional embodiment, the material information to be processed is segmented and filtered in the above step 102 to obtain a dictionary to be used, including:

[0088] Identify and process the song information to be processed to obtain initial lyrics text information;

[0089] The initial lyrics text information is segmented and filtered to obtain a dictionary to be used.

[0090] Optionally, the above-mentioned song information to be processed includes several songs to be processed.

[0091] Optionally, the above-mentioned initial lyrics text information includes several initial lyrics texts.

[0092] In this optional embodiment, as an optional implementation, the specific method of performing identification processing on the song information to be processed to obtain the initial lyrics text information is:

[0093] For any song to be processed, determine whether the song to be processed has lyrics, and obtain a lyrics determination result;

[0094] When the result of the lyrics judgment is yes, the lyrics of the song to be processed are extracted to obtain the initial lyrics text corresponding to the song to be processed;

[0095] When the above lyrics judgment result is no, the song to be processed is recognized and processed using a preset speech recognition model to obtain an initial lyrics text corresponding to the song to be processed.

[0096] It can be seen that the data processing method for constructing a domain dictionary described in the embodiment of the present invention can obtain a stand-by dictionary through recognition processing and segmentation and filtering processing of the song information to be processed, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and thus improve the generation efficiency of advertising lyrics and reduce production costs.

[0097] In another optional embodiment, the above segmentation and filtering process of the initial lyrics text information to obtain a dictionary to be used includes:

[0098] Performing character segmentation and line separation processing on the initial lyrics text information to obtain first intermediate lyrics text information;

[0099] Calculating and processing the first intermediate lyrics text information to obtain second intermediate lyrics text information;

[0100] The second intermediate lyrics text information is screened and processed by using a preset stop word list to obtain third intermediate lyrics text information; the third intermediate lyrics text information includes a plurality of lyrics text segments;

[0101] The third middle lyrics text information is processed by calculation, screening and sorting to obtain a dictionary to be used.

[0102] Optionally, the character segmentation of the initial lyrics text information is to convert the non-Chinese, English and numerical characters in the initial lyrics text into segmentation symbols.

[0103] Optionally, the above-mentioned separation symbol includes a space and / or an identifier, which is not limited in the embodiment of the present invention.

[0104] Optionally, the above-mentioned line separation processing of the initial lyrics text information is to separate the characters in the text according to the separation symbol.

[0105] Optionally, the above-mentioned first intermediate lyrics text information includes several first intermediate lyrics texts.

[0106] Optionally, the second intermediate lyrics text information includes a plurality of second intermediate lyrics texts.

[0107] In this optional embodiment, as an optional implementation, the specific method of calculating and processing the first intermediate lyrics text information to obtain the second intermediate lyrics text information is:

[0108] For any first intermediate lyric text, calculate the mutual information entropy of any two adjacent characters in the first intermediate lyric text to obtain entropy value information corresponding to the first intermediate lyric text; the above entropy value information includes a plurality of entropy values;

[0109] For any entropy value, determine whether the entropy value is greater than or equal to the entropy threshold, and obtain the entropy determination result;

[0110] When the above entropy judgment result is no, the characters corresponding to the entropy value are disconnected to obtain the second intermediate lyrics text corresponding to the entropy value.

[0111] Optionally, the entropy value represents the closeness of the connection between two characters. The larger the entropy value is, the greater the probability of forming a word between the two characters is.

[0112] In this optional embodiment, as another optional implementation, the specific method of using the preset stop word list to filter the second middle lyrics text information to obtain the third middle lyrics text information is:

[0113] The second intermediate lyrics text information is screened using a preset length screening condition to obtain length lyrics text information; the length lyrics text information includes a plurality of length lyrics texts;

[0114] Filtering the lyrics text information of the length by using a preset stop word list to obtain candidate lyrics text information; the candidate lyrics text information includes a plurality of candidate lyrics texts;

[0115] Performing frequency statistics on the candidate lyrics texts in the candidate lyrics text information to obtain first frequency information; the first frequency information includes a plurality of word frequencies;

[0116] The first frequency information is processed using a preset frequency threshold to obtain third intermediate lyrics text information.

[0117] Optionally, the specific method of filtering the second intermediate lyrics text information using the preset length filtering condition to obtain the length lyrics text information is:

[0118] For any second intermediate lyrics text, determine whether the second intermediate lyrics text meets a preset length screening condition, and obtain a length determination result;

[0119] When the above length judgment result is yes, it is determined that the second intermediate lyrics text is a length lyrics text.

[0120] Optionally, the above-mentioned length screening conditions include that the number of characters in the text is 2, and / or, the number of characters in the text is 3, and / or, the number of characters in the text is 4, which is not limited in this embodiment of the present invention.

[0121] Optionally, the specific method of filtering the above-mentioned length lyrics text information using a preset stop word list to obtain the candidate lyrics text information is:

[0122] For a lyric text of any length, determine whether the preset stop word list contains the lyric text of the length, and obtain the stop word determination result;

[0123] When the above stop word judgment result is no, the lyrics text of this length is determined as the candidate lyrics text.

[0124] Optionally, the stop word list includes several high-frequency terms but words without substantial meaning.

[0125] Optionally, the specific method of processing the first frequency information using the preset frequency threshold to obtain the third intermediate lyrics text information is:

[0126] For any word frequency, determine whether the word frequency is greater than a preset frequency threshold, and obtain a frequency determination result;

[0127] When the above frequency judgment result is yes, it is determined that the candidate lyrics text corresponding to the word frequency is the third middle lyrics text.

[0128] It can be seen that the data processing method for constructing a domain dictionary described in the embodiment of the present invention can obtain a dictionary to be used through comprehensive processing such as character segmentation, line segmentation, calculation, and screening and sorting of the initial lyrics text information, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and thus improve the generation efficiency of advertising lyrics and reduce production costs.

[0129] In yet another optional embodiment, the third intermediate lyrics text information is subjected to calculation, screening and sorting processing to obtain a dictionary to be used, including:

[0130] Performing information entropy calculation processing on the third intermediate lyrics text information to obtain first text entropy information;

[0131] Using a preset entropy threshold to filter the first text entropy information to obtain second text entropy information;

[0132] The second text entropy information is sorted to obtain text sequence information; the sorting is performed based on frequency information corresponding to the lyrics text segments;

[0133] The text sequence information is processed to obtain a dictionary to be used.

[0134] In this optional embodiment, as an optional implementation, the specific manner of sorting the second text entropy information to obtain the text sequence information is as follows:

[0135] Calculate the occurrence frequency of the second text entropy information in the initial lyrics text information to obtain second frequency information; the second frequency information includes a plurality of text frequencies;

[0136] The second text entropy information is sorted in descending order of text frequency to obtain text sequence information.

[0137] It can be seen that the data processing method for constructing a domain dictionary described in the embodiment of the present invention can obtain a stand-by dictionary through comprehensive processing such as information entropy calculation, screening processing and sorting processing of the third intermediate lyrics text information, which is more conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thus improving the generation efficiency of advertising lyrics and reducing production costs.

[0138] Embodiment 2

[0139] See also Figure 2 , Figure 2 is a flow chart of another data processing method for constructing a domain dictionary disclosed in an embodiment of the present invention. Figure 2 The data processing method for constructing a domain dictionary described is applied to a data processing system, such as a local server or a cloud server for data processing management for constructing a domain dictionary, and the embodiments of the present invention do not limit this. Figure 2 As shown, the data processing method for constructing a domain dictionary may include the following operations:

[0140] 201. Obtain information of materials to be processed.

[0141] 202. Segment and filter the material information to be processed to obtain a dictionary to be used.

[0142] 203. Use a preset word vector model to process the dictionary and marketing text information to obtain word vector information.

[0143] 204. Use a preset first word segmentation tool to process the dictionary to be used and the marketing text information to obtain the domain word information to be used.

[0144] 205. Convert and filter the word vector information and the to-be-used domain word information to obtain domain keyword information.

[0145] In the embodiment of the present invention, for the specific technical details and technical terminology of step 201-step 202, reference can be made to the detailed description of step 101-step 102 in embodiment 1, and the embodiment of the present invention will not be repeated here.

[0146] Optionally, the above-mentioned processing of the pending dictionary and marketing text information using the preset first word segmentation tool can be performed before processing the pending dictionary and marketing text information using the preset word vector model, or can be performed after processing the pending dictionary and marketing text information using the preset word vector model, or can be performed in parallel with processing the pending dictionary and marketing text information using the preset word vector model, which is not limited to the embodiments of the present invention.

[0147] Optionally, the above-mentioned word vector model includes a model based on FastText, and / or a model based on Golve, and / or a model based on Word2Vec, and / or a model based on Bert, and / or a model based on Wobert, which is not limited in the embodiment of the present invention.

[0148] Optionally, the first word segmentation tool includes a word segmenter based on jieba, and / or a word segmenter based on lac, and / or a word segmenter based on ltp, and / or a word segmenter based on thulac, and / or a word segmenter based on hanlp, and / or a word segmenter based on pskseg, which is not limited in the embodiment of the present invention.

[0149] In this optional implementation example, as an optional implementation mode, the specific method of using the preset first word segmentation tool to process the dictionary to be used and the marketing text information to obtain the field word information to be used is:

[0150] Using the preset first word segmentation tool to import the dictionary to be used, and performing word segmentation and part-of-speech tagging on the marketing text information to obtain text part-of-speech information;

[0151] The above text part-of-speech information is screened using the preset part-of-speech screening conditions to obtain the field word information to be used.

[0152] It can be seen that the data processing method for constructing a domain dictionary described in the embodiment of the present invention can obtain a stand-by dictionary by segmenting and filtering the material information to be processed, and then obtain the domain keyword information for constructing the domain dictionary through the comprehensive processing of the segmentation and filtering processing and the first word segmentation tool, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and then improve the generation efficiency of advertising lyrics and reduce production costs.

[0153] In an optional embodiment, in the above step 205, the word vector information and the to-be-used domain word information are converted and filtered to obtain domain keyword information, including:

[0154] The word vector information is used to convert the domain word information to obtain the domain word vector information;

[0155] Perform clustering calculation on domain word vector information to obtain word distance information;

[0156] The word distance information is grouped and filtered to obtain domain keyword information.

[0157] In this optional embodiment, as an optional implementation, the above-mentioned clustering calculation process is performed on the domain word vector information to obtain the word distance information in a specific manner as follows:

[0158] The domain word vector information is clustered using a preset clustering model, and cluster center information is determined according to a preset category quantity parameter; the cluster center information includes a number of cluster centers; the number of cluster centers is related to the category quantity parameter;

[0159] The Euclidean distance between the domain word vector information and the above cluster center information is calculated to obtain word distance information; the above word distance information includes several word distances.

[0160] In this optional embodiment, as another optional implementation, the specific method of performing grouping and screening processing on the word distance information to obtain the field keyword information is as follows:

[0161] The above word distance information is grouped into categories to obtain category distance information;

[0162] Sort the above category distance information in ascending order according to word distance to obtain distance sequence information;

[0163] The above distance sequence information is screened and processed using the preset vocabulary threshold information to obtain domain keyword information.

[0164] It can be seen that the data processing method for constructing a domain dictionary described in the embodiment of the present invention can obtain domain keyword information through comprehensive processing such as conversion processing, clustering calculation processing and group screening processing of the used domain word information, which is more conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thus improving the generation efficiency of advertising lyrics and reducing production costs.

[0165] In another optional embodiment, after segmenting and filtering the dictionary to be used to obtain the domain keyword information, the method further includes:

[0166] Processing the rhyming word list information to obtain a list of homophonic words;

[0167] The preset second word segmentation tool is used to process the rhyme list, field keyword information and song information to be processed to obtain training text information; the training text information is used to train the format restriction model; the format restriction model is used to constrain the format of the advertising lyrics.

[0168] Optionally, the above processing of the rhyme word table information is performed using the xpinyin library.

[0169] Optionally, the training text information includes lyrics text information segmented by special characters, and / or part-of-speech information, and / or rhyme mark information, which is not limited in the embodiment of the present invention.

[0170] Optionally, the special characters include spaces and / or punctuation marks, which are not limited in the embodiment of the present invention.

[0171] Optionally, the above-mentioned processing of the rhyme list, the field keyword information and the song information to be processed by using the preset second word segmentation tool is performed by using the second word segmentation tool and a self-built dictionary.

[0172] Optionally, the self-built dictionary includes industry category words, and / or brand words, and / or ingredient words, and / or efficacy words, and / or marketing words, which is not limited in the embodiment of the present invention.

[0173] Optionally, the second word segmentation tool includes a word segmenter based on jieba, and / or a word segmenter based on lac, and / or a word segmenter based on ltp, and / or a word segmenter based on thulac, and / or a word segmenter based on hanlp, and / or a word segmenter based on pskseg, which is not limited in the embodiment of the present invention.

[0174] It can be seen that the data processing method for constructing a domain dictionary described in the embodiment of the present invention can obtain training text information for training the format restriction model through comprehensive processing of rhyming word table information, domain keyword information and song information to be processed, which is more conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thereby improving the generation efficiency of advertising lyrics and reducing production costs.

[0175] Embodiment 3

[0176] See also Figure 3 , Figure 3 Schematic diagram of a data processing device for constructing a domain dictionary disclosed in an embodiment of the present invention. Figure 3 The described device can be applied to a data processing system, such as a local server or a cloud server for data processing management for building a domain dictionary, and the embodiments of the present invention are not limited thereto. Figure 3 As shown, the device may include:

[0177] The acquisition module 301 is used to acquire the material information to be processed; the material information to be processed includes marketing text information, and / or song information to be processed, and / or rhyme word table information;

[0178] The first processing module 302 is used to perform segmentation and filtering on the material information to be processed to obtain a standby dictionary; the standby dictionary is used to filter the words in the material text;

[0179] The second processing module 303 is used to segment and filter the dictionary to be used to obtain domain keyword information; the keyword information is strongly related to the advertising domain to which the material information to be processed belongs; the domain keywords are used to construct the domain dictionary.

[0180] It can be seen that implementation Figure 3 The described data processing device for constructing a domain dictionary can obtain domain keyword information for constructing a domain dictionary through comprehensive processing such as segmentation and filtering processing and segmentation and screening processing of the material information to be processed, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and thus improve the generation efficiency of advertising lyrics and reduce production costs.

[0181] In another optional embodiment, Figure 4 As shown, the first processing module 302 includes a first processing submodule 3021 and a second processing submodule 3022, wherein:

[0182] The first processing submodule 3021 is used to identify the song information to be processed and obtain initial lyrics text information;

[0183] The second processing submodule 3022 is used to segment and filter the initial lyrics text information to obtain a dictionary to be used.

[0184] It can be seen that implementation Figure 4 The described data processing device for constructing a domain dictionary can obtain a dictionary to be used through recognition processing and segmentation and filtering processing of the song information to be processed, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and thus improve the generation efficiency of advertising lyrics and reduce production costs.

[0185] In yet another optional embodiment, Figure 4 As shown, the second processing submodule 3022 performs segmentation and filtering processing on the initial lyrics text information to obtain a dictionary to be used, including:

[0186] Performing character segmentation and line separation processing on the initial lyrics text information to obtain first intermediate lyrics text information;

[0187] Calculating and processing the first intermediate lyrics text information to obtain second intermediate lyrics text information;

[0188] The second intermediate lyrics text information is screened and processed by using a preset stop word list to obtain third intermediate lyrics text information; the third intermediate lyrics text information includes a plurality of lyrics text segments;

[0189] The third middle lyrics text information is processed by calculation, screening and sorting to obtain a dictionary to be used.

[0190] It can be seen that implementation Figure 4 The described data processing device for constructing a domain dictionary can obtain a dictionary to be used through comprehensive processing such as character segmentation, line segmentation, calculation, and screening and sorting of the initial lyrics text information, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and thus improve the generation efficiency of advertising lyrics and reduce production costs.

[0191] In yet another optional embodiment, Figure 4 As shown, the second processing submodule 3022 performs calculation, screening and sorting processing on the third middle lyrics text information to obtain a dictionary to be used, including:

[0192] Performing information entropy calculation processing on the third intermediate lyrics text information to obtain first text entropy information;

[0193] Using a preset entropy threshold to filter the first text entropy information to obtain second text entropy information;

[0194] The second text entropy information is sorted to obtain text sequence information; the sorting is performed based on frequency information corresponding to the lyrics text segments;

[0195] The text sequence information is processed to obtain a dictionary to be used.

[0196] It can be seen that implementation Figure 4 The described data processing device for constructing a domain dictionary can obtain a dictionary to be used through comprehensive processing such as information entropy calculation, screening processing and sorting processing of the third intermediate lyrics text information, which is more conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thus improving the generation efficiency of advertising lyrics and reducing production costs.

[0197] In yet another optional embodiment, Figure 4 As shown, the second processing module 303 performs segmentation and screening processing on the dictionary to be used, and obtains the field keyword information in the following specific manner:

[0198] Use the preset word vector model to process the dictionary and marketing text information to obtain word vector information;

[0199] Using the preset first word segmentation tool to process the dictionary to be used and the marketing text information to obtain the domain word information to be used;

[0200] The word vector information and the to-be-used domain word information are converted and filtered to obtain domain keyword information.

[0201] It can be seen that implementation Figure 4 The described data processing device for constructing a domain dictionary can obtain a stand-by dictionary by segmenting and filtering the information of the material to be processed, and then obtain domain keyword information for constructing the domain dictionary through comprehensive processing of the segmentation and filtering processing and the first word segmentation tool, which is beneficial to reduce the requirements for song materials, improve the processing efficiency of song materials, and thereby improve the generation efficiency of advertising lyrics and reduce production costs.

[0202] In yet another optional embodiment, Figure 4 As shown, the second processing module 303 performs conversion and screening processing on the word vector information and the to-be-used domain word information to obtain the domain keyword information in the following specific manner:

[0203] The word vector information is used to convert the domain word information to obtain the domain word vector information;

[0204] Perform clustering calculation on domain word vector information to obtain word distance information;

[0205] The word distance information is grouped and filtered to obtain domain keyword information.

[0206] It can be seen that implementation Figure 4 The described data processing device for constructing a domain dictionary can obtain domain keyword information through comprehensive processing such as conversion processing, clustering calculation processing and group screening processing of the used domain word information, which is more conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thus improving the generation efficiency of advertising lyrics and reducing production costs.

[0207] In yet another optional embodiment, Figure 4 As shown, after the second processing module 303 performs segmentation and screening processing on the dictionary to be used to obtain the domain keyword information, the device further includes:

[0208] The third processing module 304 is used to process the rhyming word list information to obtain a homophonic word list;

[0209] The fourth processing module 305 is used to use the preset second word segmentation tool to process the rhyme list, field keyword information and song information to be processed to obtain training text information; the training text information is used to train the format restriction model; the format restriction model is used to constrain the format of the advertising lyrics.

[0210] It can be seen that implementation Figure 4The described data processing device for constructing a domain dictionary can obtain training text information for training a format restriction model by comprehensive processing of rhyming word table information, domain keyword information and song information to be processed, which is more conducive to reducing the requirements for song materials, improving the processing efficiency of song materials, and thereby improving the generation efficiency of advertising lyrics and reducing production costs.

[0211] Embodiment 4

[0212] See also Figure 5 , Figure 5 is a structural diagram of another data processing device for constructing a domain dictionary disclosed in an embodiment of the present invention. Figure 5 The described device can be applied to a data processing system, such as a local server or a cloud server for data processing management for building a domain dictionary, and the embodiments of the present invention are not limited thereto. Figure 5 As shown, the device may include:

[0213] A memory 401 storing executable program codes;

[0214] a processor 402 coupled to the memory 401;

[0215] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the data processing method for constructing a domain dictionary described in the first or second embodiment.

[0216] Embodiment 5

[0217] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the data processing method for constructing a domain dictionary described in the first or second embodiment.

[0218] Embodiment 6

[0219] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps in the data processing method for constructing a domain dictionary described in Embodiment 1 or Embodiment 2.

[0220] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without paying creative labor.

[0221] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0222] Finally, it should be noted that the data processing method and device for constructing a domain dictionary disclosed in the embodiment of the present invention discloses only the preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data processing method for constructing a domain dictionary, characterized in that: The method comprises: Acquire information of material to be processed; the material information to be processed includes marketing text information, and / or song information to be processed, and / or rhyme word list information; Segment and filter the material information to be processed to obtain a dictionary to be used; the dictionary to be used is used to filter words in the material text; The dictionary to be used is segmented and screened to obtain domain keyword information; the keyword information is strongly related to the advertising domain to which the material information to be processed belongs; the domain keywords are used to construct a domain dictionary; The segmentation and screening of the dictionary to be used to obtain domain keyword information includes: Using a preset word vector model to process the dictionary to be used and the marketing text information to obtain word vector information; Using a preset first word segmentation tool to process the dictionary to be used and the marketing text information to obtain field word information to be used; Using the word vector information, the to-be-used domain word information is converted to obtain domain word vector information; Performing clustering calculation on the domain word vector information to obtain word distance information; The word distance information is grouped and screened to obtain domain keyword information.

2. The data processing method for constructing a domain dictionary according to claim 1, characterized in that: The segmenting and filtering of the to-be-processed material information to obtain a dictionary to be used includes: Performing identification processing on the song information to be processed to obtain initial lyrics text information; The initial lyrics text information is segmented and filtered to obtain a dictionary to be used.

3. The data processing method for constructing a domain dictionary according to claim 2, characterized in that: The segmentation and filtering of the initial lyrics text information to obtain a dictionary to be used includes: Performing character segmentation and line separation processing on the initial lyrics text information to obtain first intermediate lyrics text information; Performing calculation processing on the first intermediate lyrics text information to obtain second intermediate lyrics text information; The second intermediate lyrics text information is screened and processed by using a preset stop word list to obtain third intermediate lyrics text information; the third intermediate lyrics text information includes a plurality of lyrics text segments; Performing calculation, screening and sorting processing on the third intermediate lyrics text information to obtain a dictionary to be used; The step of calculating and processing the first intermediate lyrics text information to obtain the second intermediate lyrics text information includes: For any of the first intermediate lyrics texts, calculate the mutual information entropy of any two adjacent characters in the first intermediate lyrics text to obtain the entropy value information corresponding to the first intermediate lyrics text; the entropy value information includes a plurality of entropy values; for any of the entropy values, determine whether the entropy value is greater than or equal to an entropy threshold to obtain an entropy determination result; when the entropy determination result is no, disconnect the characters corresponding to the entropy value to obtain the second intermediate lyrics text corresponding to the entropy value.

4. The data processing method for constructing a domain dictionary according to claim 3, characterized in that: The step of calculating, screening and sorting the third intermediate lyrics text information to obtain a dictionary to be used includes: Performing information entropy calculation processing on the third intermediate lyrics text information to obtain first text entropy information; Using a preset entropy threshold to filter the first text entropy information to obtain second text entropy information; Sorting the second text entropy information to obtain text sequence information; the sorting is performed based on frequency information corresponding to the lyrics text segments; The text sequence information is processed to obtain a dictionary to be used.

5. The data processing method for constructing a domain dictionary according to claim 1, characterized in that: After the segmentation and screening process is performed on the dictionary to be used to obtain the domain keyword information, the method further includes: Processing the rhyming word list information to obtain a homophonic word list; The preset second word segmentation tool is used to process the homophonic list, the field keyword information and the song information to be processed to obtain training text information; the training text information is used to train the format restriction model; the format restriction model is used to constrain the format of the advertisement lyrics.

6. A data processing device for constructing a domain dictionary, characterized in that: The device comprises: An acquisition module, used for acquiring material information to be processed; the material information to be processed includes marketing text information, and / or song information to be processed, and / or rhyme word table information; The first processing module is used to segment and filter the material information to be processed to obtain a dictionary to be used; the dictionary to be used is used to filter words in the material text; The second processing module is used to use a preset word vector model to process the standby dictionary and the marketing text information to obtain word vector information, use a preset first word segmentation tool to process the standby dictionary and the marketing text information to obtain standby domain word information, use the word vector information to convert the standby domain word information to obtain domain word vector information, perform clustering calculation on the domain word vector information to obtain word distance information, and perform grouping and screening on the word distance information to obtain domain keyword information; the keyword information is strongly related to the advertising field to which the material information to be processed belongs; the domain keywords are used to construct a domain dictionary.

7. A data processing device for constructing a domain dictionary, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the data processing method for constructing a domain dictionary according to any one of claims 1 to 5.

8. A computer storable medium, characterized in that: The computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the data processing method for constructing a domain dictionary according to any one of claims 1 to 5.

Images