Method, device, electronic device and storage medium for estimating text reading time

Abstract

The present disclosure provides a method, device, electronic device and storage medium for estimating text processing time, the method comprising: determining the text content and each sentence contained in the target text to be processed; calculating the information volume of the target text based on the text content and each sentence; obtaining the information volume processing rate of the target object; estimating the processing time of the target object for the target text based on the information volume of the target text and the information volume processing rate. The embodiments of the present disclosure can improve the accuracy of estimating text processing time.

Description

Technical Field

[0001] The present disclosure relates to the field of natural language processing, and in particular to a method, device, electronic device and storage medium for estimating text reading time. Background Art

[0002] In today's digital age, people often read digital texts in their daily lives. Estimating the reading time of users and then providing them with a forecast is an important part of improving users' digital reading experience.

[0003] In the prior art, when estimating the reading time of a text, the length of the text is divided by the reading speed to calculate the reading time. However, the reading time required for texts of the same length may be very different: under normal circumstances, for a paper and a novel of the same length, the reading time required for a user to read the paper is significantly longer than the reading time required to read the novel. However, when the prior art estimates the reading time of the user for the two, the result will be: the reading time required for the user to read the paper is equal to the reading time required to read the novel. Obviously, the prior art estimates the reading time of the text inaccurately. Summary of the invention

[0004] An object of the present disclosure is to provide a method, device, electronic device and storage medium for estimating text reading time, so as to improve the accuracy of estimating the reading time of a user for a text.

[0005] According to one aspect of an embodiment of the present disclosure, a method for estimating text reading time is disclosed, the method comprising:

[0006] Determine the text content and each sentence contained in the target text to be processed;

[0007] Calculating the information content of the target text based on the text content and the various sentences;

[0008] Get the user's information reading rate;

[0009] Based on the information volume of the target text and the information reading rate, the reading time of the target text by the user is estimated.

[0010] According to one aspect of an embodiment of the present disclosure, a device for estimating text reading time is disclosed, the device comprising:

[0011] A determination module, used to determine the text content and each sentence contained in the target text to be processed;

[0012] An information volume calculation module, used for calculating the information volume of the target text based on the text content and each sentence;

[0013] An acquisition module is used to acquire the user's information reading rate;

[0014] The estimation module is used to estimate the reading time of the target text by the user based on the information volume of the target text and the reading speed of the information volume.

[0015] According to one aspect of an embodiment of the present disclosure, an electronic device for estimating reading time is disclosed, comprising: a memory storing computer-readable instructions; and a processor reading the computer-readable instructions stored in the memory to execute any one of the methods described in the above claims.

[0016] According to one aspect of an embodiment of the present disclosure, a computer program medium is disclosed, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor of a computer, the computer is caused to execute the method described in any one of the above claims.

[0017] The technical solution of the disclosed embodiment calculates the amount of information of the text based on the text content and each sentence contained in the text when estimating the reading time of the text, thereby being able to measure the amount of information of the text more three-dimensionally and accurately, thereby improving the accuracy of estimating the user's reading time for the text.

[0018] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other objects, features and advantages of the present disclosure will become more apparent by describing in detail example embodiments thereof with reference to the attached drawings.

[0021] Figure 1A A system architecture diagram in a specific application scenario according to an embodiment of the present disclosure is shown.

[0022] Figure 1B A system architecture diagram in a specific application scenario according to an embodiment of the present disclosure is shown.

[0023] Figure 2A The following is a diagram showing an interface of a user terminal according to an embodiment of the prior art.

[0024] Figure 2B A diagram showing an interface of a user terminal according to an embodiment of the present disclosure is shown.

[0025] Figure 3 A flow chart of a method for estimating text reading time according to an embodiment of the present disclosure is shown.

[0026] Figure 4 A flow chart of a method for calculating the amount of information of a target text based on the text content and each sentence according to an embodiment of the present disclosure is shown.

[0027] Figure 5A A graph showing the relationship between text length and information volume according to an embodiment of the present disclosure is shown.

[0028] Figure 5B A graph showing the relationship between text length and information volume according to an embodiment of the present disclosure is shown.

[0029] Figure 6 A flow chart of a method for calculating the amount of information of a target text based on the text content and each sentence according to an embodiment of the present disclosure is shown.

[0030] Figure 7 A module diagram of a device for estimating text reading time according to an embodiment of the present disclosure is shown.

[0031] Figure 8 A hardware structure diagram for estimating text reading time according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0032] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete and the concepts of the example embodiments will be fully conveyed to those skilled in the art. The accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0033] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0034] Some of the blocks shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0035] First reference Figure 1A , 1B Describe the performance of the embodiments of the present disclosure in specific application scenarios.

[0036] Figure 1A The system architecture composition of an embodiment of the present disclosure in a specific application scenario is shown: a subscription account server 10 and a subscription user terminal 20. Among them, the subscription account server 10 can push subscription account articles to the subscription user terminal 20, and the user can read the subscription account articles on the subscription user terminal 20. Specifically, when the user enters the subscription account on the subscription user terminal 20, the subscription user terminal 20 displays the titles of each subscription account article in sequence on the display interface. At this time, the subscription account server 10 can estimate the reading time of each subscription account article for the user, and transmit the estimated reading time to the subscription user terminal 20, so that the subscription user terminal 20 can display the estimated reading time under the title of the corresponding subscription account article to provide a forecast to the user.

[0037] Figure 1B The system architecture of an embodiment of the present disclosure in another application scenario is shown: an electronic library server 10 and a reader terminal 20. The reader can request to read an electronic journal from the electronic library server 10 through the reader terminal 20. After receiving the request, the electronic library 10 can estimate the reading time of the electronic journal when downloading the corresponding electronic journal to the corresponding reader terminal 20, and also download the estimated reading time to the reader terminal 20. When the reader terminal 20 provides the electronic journal to the reader for reading, the estimated reading time can be displayed at a predetermined position on the cover of the electronic journal to inform the reader.

[0038] It should be noted that Figure 1A , Figure 1B The possible application scenarios and corresponding system architecture components of the present disclosure are only shown by way of example. This does not mean that the present disclosure can only be applicable to the two application scenarios with the shown system architecture components, nor does it mean that the executor of the estimated text reading time of the present disclosure can only be the server.

[0039] Reference below Figure 2A , 2B The following are respectively described in the same specific application scenario: a user interface for estimating text reading time in the prior art and a user interface for estimating text reading time in the embodiment of the present disclosure, so as to show the general performance of the embodiment of the present disclosure in actual application and compare it with the prior art.

[0040] Figure 2AThe user interface of an embodiment of the prior art in a specific application scenario is shown: In this embodiment, the user opens the subscription account message on the user end and wants to read the articles pushed by each subscription account. Before the user clicks on the target article to read, the execution subject of the prior art estimates the time it takes the user to read the target article and displays the estimated reading time under the title of the corresponding target article.

[0041] like Figure 2A As shown, in this embodiment, the estimated user reading time for the target article is linearly proportional to the number of words in the target article: the target article "Where to go for fun during the Mid-Autumn Festival" has 1000 words, and the estimated user reading time is 1 minute; the target article "Popular Science: The Development History of Carbonated Beverages" has 3000 words, and the estimated user reading time is 3 minutes; the target article "A Brief Textual Research and Introduction to Tang Dynasty Costumes" has 3000 words, and the estimated user reading time is 3 minutes.

[0042] Figure 2B The user interface of an embodiment of the present disclosure in the same application scenario is shown: In this embodiment, the user opens the subscription account message on the user end and wants to read the articles pushed by each subscription account. Before the user clicks on the target article to read, the execution subject of the present disclosure estimates the time it takes the user to read the target article and displays the estimated reading time under the title of the corresponding target article.

[0043] like Figure 2B As shown, in this embodiment, the estimated user reading time for the target article is not linearly proportional to the number of words in the target article: the target article "Where to go for fun during the Mid-Autumn Festival" has 1000 words, and the estimated user reading time is 1 minute; the target article "Popular Science: The Development History of Carbonated Beverages" has 3000 words, and the estimated user reading time is 6 minutes; the target article "A Brief Textual Research and Introduction to Tang Dynasty Costumes" has 3000 words, and the estimated user reading time is 10 minutes.

[0044] It is understandable that, considering the differences in text content, even if the number of words is the same, the actual amount of information contained in the entertainment content text and the scientific research content text of the same number of words is different. Compared with the entertainment content text, the scientific research content text of the same number of words will contain more information, so that users need to spend more time reading the text content to extract and absorb the information. It can be seen that the existing technology simply measures the amount of information contained in the text based on the text length as the standard, and cannot accurately estimate the text reading time.

[0045] Specific, combined with common sense and Figure 2A , Figure 2BFrom the application scenario shown, it can be understood that: even with the same number of words, the article "A Brief Textual Research and Introduction to Tang Dynasty Costumes" contains more information than the article "Popular Science: The Development History of Carbonated Beverages", and the article "Popular Science: The Development History of Carbonated Beverages" contains more information than the article "Where to Go for Fun During the Mid-Autumn Festival". That is to say, even with the same number of words, users need to spend more time reading the article "A Brief Textual Research and Introduction to Tang Dynasty Costumes" than reading the article "Popular Science: The Development History of Carbonated Beverages", and it takes more time to read the article "Popular Science: The Development History of Carbonated Beverages" than reading the text "Where to Go for Fun During the Mid-Autumn Festival". Obviously, in the prior art, the amount of information contained in the text is measured only by the text length of the text, which will result in the estimated text reading time being linearly proportional to the number of words, such as Figure 2A which is clearly inaccurate.

[0046] A more accurate and realistic answer would be: Figure 2B The disclosed estimation of the text reading time is shown as follows: for texts containing more information and requiring users to spend more time to read, the estimated reading time is longer, rather than simply being linearly proportional to the number of words in the text.

[0047] The specific implementation process of the embodiment of the present disclosure is described in detail below.

[0048] First, regarding the execution subject of the embodiment of the present disclosure, the execution subject of the embodiment of the present disclosure can be a server or a user end.

[0049] In one embodiment, the execution subject of estimating the text reading time is a server. In this embodiment, when the estimation of the user's reading time for the target text is triggered, the server determines the text content and each sentence contained in the target text, and then calculates the information volume of the target text, obtains the user's information volume reading rate, and then estimates the user's reading time for the target text.

[0050] Furthermore, in combination with the user's operation on the user side, the general process example of this embodiment is as follows: the user clicks on the user side to enter the text reading interface, triggering the server to estimate the user's reading time for the target text. Among them, the server determines each text displayed on the text reading interface as a target text of the user in turn, and estimates the user's reading time for each target text in turn. The server determines the text content and each sentence contained in the target text, and then calculates the amount of information in the target text, obtains the user's information reading rate, and then estimates the user's reading time for the target text, and sends the estimated reading time to the user side, so that the user side displays the estimated reading time at a predetermined position of the target text title in the text reading interface.

[0051] The advantage of this embodiment is that, since the computing power of the server is more powerful than that of the user end, the server can estimate the reading time more quickly, thereby improving the user's reading experience.

[0052] In one embodiment, the execution subject of estimating the text reading time is a user terminal. In this embodiment, when the user's reading time for the target text is estimated, the user terminal determines the text content and each sentence contained in the target text, and then calculates the information volume of the target text, obtains the user's information volume reading rate, and then estimates the user's reading time for the target text.

[0053] Furthermore, in combination with the user's operation on the user terminal, the general process of this embodiment is exemplified as follows: the user clicks on the user terminal to enter the text reading interface, triggering the user terminal to estimate the user's reading time for the target text. Among them, the user terminal successively determines each text displayed on the text reading interface as a target text of the user, and successively estimates the user's reading time for each target text. The user terminal determines the text content and each sentence contained in the target text, and then calculates the amount of information in the target text, obtains the user's information reading rate, and then estimates the user's reading time for the target text, and displays the estimated reading time at a predetermined position of the target text title in the text reading interface.

[0054] The advantage of this embodiment is that the reading time is estimated by the user terminal, so that the entire text reading time estimation process does not need to rely on the server, and even if the user terminal is offline, the reading time estimation service can be provided to the user.

[0055] It should be noted that the above embodiments of the execution subject of the present disclosure are all exemplary descriptions and should not limit the function and scope of use of the present disclosure.

[0056] like Figure 3 As shown, a method for estimating text reading time includes:

[0057] Step 310: determining the text content and each sentence contained in the target text to be processed;

[0058] Step 320: Calculate the information content of the target text based on the text content and each sentence;

[0059] Step 330: Obtain the user's information reading rate;

[0060] Step 340: based on the information volume of the target text and the information reading rate, estimate the user's reading time for the target text.

[0061] Target text refers to text that can be regarded as the target object of user reading.

[0062] The information volume of the target text is used to measure the amount of information contained in the target text. It should be noted that the information volume in the prior art is self-defined and takes the text length as the dimension of the information volume contained in the target text, and is not the information volume calculated based on Shannon's information theory as strictly defined in the field of communications. The information volume of the present disclosure is also a self-defined dimension. The difference from the prior art is that the present disclosure calculates the information volume of the target text based on the text content and each sentence, which can express the information volume of the target text in a more three-dimensional way, rather than simply calculating the information volume of the target text based on the text length.

[0063] In one embodiment, the target text is each text displayed on the text reading interface of the user end.

[0064] In this embodiment, each text in the user-side text reading interface is a target text. The execution subject of the present disclosure estimates the user's reading time for each target text in turn according to the method provided by the present disclosure.

[0065] The advantage of this embodiment is that it automatically estimates the reading time of each text in the user-side text reading interface, and can display the estimated reading time of multiple texts to the user in batches.

[0066] In one embodiment, the target text is the text indicated by the user's reading instruction.

[0067] In this embodiment, after receiving a user's instruction to read a text, the execution subject of the present disclosure uses the text indicated by the reading instruction as the target text, and estimates the user's reading time for the target text according to the method provided by the present disclosure.

[0068] The advantage of this embodiment is that it only estimates the reading time of the text indicated by the user, thereby reducing the amount of calculation and improving the utilization rate of calculation resources.

[0069] In one embodiment, the reading instruction is generated in response to a predetermined gesture of the user on the user terminal to the text, wherein the predetermined gesture to the text includes but is not limited to: clicking on the text title, swiping left on the text title.

[0070] In one embodiment, the reading instruction is a voice instruction of the user to the text on the user terminal. In this embodiment, the user can indicate the target text by issuing a voice instruction to the user terminal.

[0071] It should be noted that the above embodiments of the reading instructions are merely exemplary and should not limit the functions and scope of use of the present disclosure.

[0072] In the embodiment of the present disclosure, the text content and each sentence contained in the target text are determined, and the information volume of the target text is calculated on this basis. The user's information volume reading rate is obtained, and then based on the information volume of the target text and the user's information volume reading rate, the user's reading time for the target text is estimated.

[0073] The specific process of each step of the present disclosure is described in detail below.

[0074] Step 310: Determine the text content and each sentence contained in the target text to be processed.

[0075] In one embodiment, the step of determining the text content and each sentence contained in the target text to be processed includes:

[0076] Determining the text content contained in the target text;

[0077] The text content is divided into sentences to obtain sentence division results of the text content.

[0078] In this embodiment, after determining the target text to be processed, the text content contained in the target text is extracted and determined. The text content is sentence-separated to obtain the sentence-separation result of the text content, and then the various sentences contained in the target text can be determined. Specifically, the number of sentences contained in the target text and the content of each sentence can be determined from the sentence-separation result of the text content. Among them, the text content can be sentence-separated using the existing sentence-separation method, so the specific process of sentence-separating the text content will not be repeated here.

[0079] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0080] The following describes the specific process of calculating the information content of the target text based on the text content and each sentence.

[0081] Step 320: Calculate the information content of the target text based on the text content and the various sentences.

[0082] In one embodiment, if Figure 4 As shown, the information amount of the target text is calculated based on the text content and each sentence, including:

[0083] Step 3201: Obtain a word vector for each word in the text content;

[0084] Step 3202: based on the word vector of each word in the text content, the words contained in each sentence and a preset convex hull algorithm, respectively calculate the convex hull volume of each sentence;

[0085] Step 3203: Calculate the sum of the convex hull volumes of the sentences, and determine the amount of information of the target text according to the sum of the convex hull volumes.

[0086] Convex hull, or Convex Hull, refers to a real vector space V, for a given set X, the intersection S of all convex sets containing X is called the convex hull of set X. In a two-dimensional Euclidean space, the convex hull can be imagined as a rubber band that just wraps all points. Further, extended to multi-dimensional space, the convex hull can be imagined as a convex body that just surrounds all points. Specifically in this embodiment, each sentence contained in the target text can be regarded as a set, and each word in the sentence can be regarded as a point in the set, then each word in the sentence can together form a convex hull. The volume of the convex hull is used to measure the amount of information of the corresponding sentence, and then the amount of information of the target text is determined.

[0087] Word vectors refer to words represented in the form of vectors. Different words have different corresponding word vectors, and word vectors locate the unique position of the corresponding word in the vector space. Therefore, word vectors can be used to determine the distance between each word in the vector space, and then the volume of the convex hull formed by each word in the vector space can be determined, and then the amount of information in the target text can be determined.

[0088] When a user reads a text, he does not read the entire text before extracting and absorbing the information therein. It is understandable that when a user reads a text, he pauses briefly after reading each sentence to extract and absorb the information contained in the sentence. Therefore, in this embodiment, each sentence is taken as a convex hull, and the convex hull volume of each sentence is calculated based on the word vector of each word, and then the information content of the target text is determined according to the sum of the convex hull volumes of each sentence.

[0089] In this embodiment, the word vector of each word in the text content contained in the target text is obtained. For each sentence in the text content, the convex hull volume of each sentence is calculated based on the word vector of each word, the words contained in each sentence and the preset convex hull algorithm. The information content of the target text is determined according to the sum of the convex hull volumes of each sentence.

[0090] The advantage of this embodiment is that the word vector can more accurately measure the similarity in meaning between words. Furthermore, on the basis of the word vector, the information amount of the sentence is measured by the convex hull volume, and the information amount of the text is measured by the sum of the convex hull volumes, so that the measurement of the information amount is more three-dimensional and accurate.

[0091] In one embodiment, obtaining a word vector for each word in the text content includes:

[0092] Performing word segmentation on the text content to obtain a word segmentation result of the text content;

[0093] The word segmentation result is processed based on a pre-trained word vector model to obtain a word vector for each word in the text content.

[0094] In this embodiment, after determining the text content and each sentence contained in the target text, the text content is segmented to obtain the corresponding segmentation result. The segmentation result is then processed based on the pre-trained word vector model to obtain the word vector of each word in the text content.

[0095] In one embodiment, the word vector model is a BERT model, i.e., a Bidirectional Encoder Representations from Transformers model, which is a representation method of a pre-trained language open sourced by Google. Through the BERT model, the word vectors of each word in the text content can be obtained. Specifically, the BERT model is used to embed each word, and the last four layers are superimposed to obtain the word vector of each word.

[0096] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0097] In one embodiment, the words included in each sentence are determined by the following method: based on the sentence segmentation results and word segmentation results of the text content, the words included in each sentence are determined.

[0098] In this embodiment, by segmenting the text content into sentences and words, each sentence and each word included in each sentence can be determined according to the sentence segmentation results and word segmentation results.

[0099] Furthermore, after obtaining the word vector of each word in the text content and determining each word contained in each sentence, the convex hull volume of each sentence can be calculated using a preset convex hull algorithm. The convex hull algorithm used to calculate the convex hull volume of each sentence is a prior art, so the specific content of the convex hull algorithm will not be described here.

[0100] In one embodiment, determining the amount of information of the target text according to the sum of the convex hull volumes includes:

[0101] The sum of the convex hull volumes of the individual sentences is determined as the information content of the target text.

[0102] It is understandable that the process of the user acquiring the information amount of the target text can be regarded as being averaged for each part of the text content contained in the target text. That is, it is considered that when the user reads the target text, whether it is the beginning, the middle, or the end of the text content contained in the target text, they all read at the same rate. Therefore, in this embodiment, the convex hull volumes of each sentence are directly added, and the sum of the obtained convex hull volumes is determined as the information amount of the target text.

[0103] For example, if the target text contains 5 sentences, and the convex hull volumes of these 5 sentences are 100 units, 200 units, 150 units, 100 units, and 200 units, respectively, then the amount of information in the target text is (100+200+150+100+200)=750 units.

[0104] Specifically, in this embodiment, the information amount of the target text can be calculated by the following formula:

[0105]

[0106] Where I represents the amount of information in the target text; i represents the order of the sentence; n represents the number of sentences in the target text; V hull represents the equation for calculating the volume of the convex hull; f hull represents the equation for calculating the convex hull; e represents the word vector of the word; Emb(e) represents the equation for calculating the word vector of the word; sentence i Represents the i-th sentence in the target text. Among them, V hull and f hull Together they constitute the mathematical representation of the above-mentioned preset convex hull algorithm; Emb(e) is the mathematical representation of the above-mentioned pre-trained word vector model.

[0107] The advantage of this embodiment is that the information content of the target text is calculated by directly adding the sum of the convex hull volumes of each sentence, and the calculation process is simple and fast.

[0108] In one embodiment, determining the amount of information of the target text according to the sum of the convex hull volumes includes:

[0109] Based on the preset weights for the respective sentences, the weighted sum of the convex hull volumes of the respective sentences is determined as the information amount of the target text.

[0110] It is understandable that, although the amount of information of the target text is generally evenly distributed in the text content it contains, the process of obtaining the amount of information of the target text by the user may not be even for each part of the text content contained in the target text. Further, due to the reading habits of the public, they tend to read the beginning and end of the text carefully, and read the middle of the text roughly. Therefore, when the user reads the target text, the user's acquisition rate of the amount of information at the beginning and end of the text content contained in the target text is higher than that of the middle of the text content contained in the target text. For example: the beginning and end of the text content contained in the target text contain 1000 units of information, and the middle contains 2000 units of information. Since the user reads the beginning and the end more carefully, the user actually obtains 900 units of information at the beginning and the end (corresponding to a 90% acquisition rate), and the actual amount of information obtained in the middle is 1600 units of information (corresponding to a 80% acquisition rate).

[0111] Compared with the amount of information contained in the text itself, the amount of information actually obtained by the user is the factor that more directly affects the user's reading time. Therefore, in this embodiment, for each sentence of the target text, a corresponding weight is preset, and the weight sum of the convex hull volume of each sentence is determined as the amount of information of the target text, so as to more accurately estimate the user's reading time for the target text on this basis.

[0112] In one embodiment, weights are preset for each sentence, including: dividing the text content included in the target text into regions according to preset regions and corresponding weight division rules, and assigning corresponding weights to the sentences included in each region.

[0113] In this embodiment, after determining the text content and each sentence included in the target text, the text content is divided into regions according to preset regions and corresponding weight division rules, and corresponding weights are respectively assigned to the sentences included in each region.

[0114] For example: the preset area and corresponding weight division rules are: divide the first 20% of the text content into the beginning area, and assign a corresponding weight of 0.9 to each sentence contained therein; divide the last 20% of the text content into the ending area, and assign a corresponding weight of 0.9 to each sentence contained therein; divide the rest of the text content into the middle area, and assign a weight of 0.8 to each sentence contained therein.

[0115] If the target text contains 5 sentences, the convex hull volumes of these 5 sentences are 100 units, 200 units, 150 units, 100 units, and 200 units respectively. Then the amount of information of the target text is (100×0.9+200×0.8+150×0.8+100×0.8+200×0.9)=630 units.

[0116] The advantage of this embodiment is that the amount of information of the target text calculated on the basis of dividing the text content into regions and assigning weights is closer to the actual amount of information obtained by the user on the target text.

[0117] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0118] Another specific process of calculating the information content of the target text based on the text content and each sentence is described below.

[0119] In one embodiment, the calculating the information amount of the target text based on the text content and the various sentences includes:

[0120] Determining the text length of the text content and the number of sentences included in the target text according to the text content and the sentences;

[0121] The product of the text length and the number of sentences is calculated, and the information volume of the target text is determined according to the product of the text length and the number of sentences.

[0122] Specifically, after a large number of experiments, it was found that the amount of information determined by the sum of the convex hull volumes based on the word vector of each word has a certain correlation with the product of the text length and the number of sentences. More specifically, the amount of information determined by the sum of the convex hull volumes based on the word vector of each word is similar to the distribution of the information determined by the square root of the product of the text length and the number of sentences.

[0123] like Figure 5A , Figure 5B As shown, Figure 5A The relationship between the amount of information determined based on the word vector of each word and the sum of the convex hull volumes and the length of the text is shown; Figure 5B The figure shows the relationship between the amount of information determined by the product of the text length and the number of sentences and the product of the text length and the number of sentences. Specifically, Figure 5B A graph showing the relationship between the amount of information determined based on the square root of the product of the text length and the number of sentences and the square root of the product of the text length and the number of sentences is shown.

[0124] Among them, L represents the length of the text; n represents the number of sentences; 1e7 represents 107 , used to indicate the magnitude of coordinate numbers. Specifically, Figure 5A Middle: The horizontal axis is the length of the text, and the vertical axis is the amount of information; Figure 5B Middle: The horizontal axis is the square root of the product of the text length and the number of sentences, and the vertical axis is the amount of information.

[0125] Figure 5A The distribution of the points in is approximately a straight line with a positive slope. Figure 5B The distribution of each point in is also similar to a straight line with a positive slope. It can be seen that Figure 5B The expression effect of information quantity and Figure 5A The effect of expressing the amount of information is very similar.

[0126] Therefore, this embodiment provides a method that can achieve a similar effect as the above method based on the word vector of each word and then determining the amount of information based on the sum of the convex hull volumes, and can more quickly calculate the amount of information of the target text. Specifically, in this embodiment, the text length of the text content contained in the target text and the number of sentences contained in the target text are determined. The product of the text length and the number of sentences is calculated, and then the amount of information of the target text is determined based on the product of the text length and the number of sentences.

[0127] In one embodiment, determining the amount of information of the target text according to the product of the text length and the number of sentences includes:

[0128] The information content of the target text is obtained by multiplying the square root of the product of the text length and the number of sentences by a preset positive number.

[0129] In this embodiment, the information amount determined based on the word vector of each word and the sum of the convex hull volumes is similar to the distribution of the information amount determined by the square root of the product of the text length and the number of sentences. More specifically, it can be seen that the information amount of the target text is approximately proportional to the square root of the product of its corresponding text length and the number of sentences. Therefore, the square root of the product of the text length and the number of sentences is multiplied by the preset formal to obtain the information amount of the target text.

[0130] Specifically, in this embodiment, the information amount of the target text can be calculated by the following formula:

[0131]

[0132] Among them, n represents the number of sentences, L represents the text length, and I represents the information content of the target text. From this formula, it can be seen that the information content of the target text is proportional to the square root of the product of the text length and the number of sentences. Therefore, a positive number can be preset and multiplied by the square root of the product of the text length and the number of sentences to obtain the information content of the target text.

[0133] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0134] Another specific process of calculating the information content of the target text based on the text content and each sentence is described below.

[0135] In one embodiment, if Figure 6 As shown, the information amount of the target text is calculated based on the text content and each sentence, including:

[0136] Step 3201', based on the text content and the various sentences, determine the information density of the target text and the text length of the text content;

[0137] Step 3202': calculate the product of the information density and the text length, and determine the information volume of the target text according to the product of the information density and the text length.

[0138] In this embodiment, in order to reduce the amount of calculation, the information density of the target text and the text length of the text content can be determined based on the text content and each sentence, and then the information volume of the target text can be determined according to the product of the information density and the text length. Among them, since the information density of the target text can be calculated by sampling the target text and then calculating the information density of the sampled part of the text, the calculation of the convex hull volume of each sentence of the target text is avoided, thereby reducing the amount of calculation.

[0139] In one embodiment, determining the information density of the target text based on the text content and the individual sentences includes:

[0140] Determining the text content and sentences included in the specified segment of the target text according to the text content and the sentences;

[0141] Calculate the information content of the specified segment based on the text content and sentences contained in the specified segment;

[0142] The ratio between the information amount of the designated segment and the text length of the text content contained in the designated segment is calculated, and the information density of the target text is determined according to the ratio.

[0143] It can be understood that the amount of information contained in the same text is roughly evenly distributed, and even if it is located in different segments of the target text, the information density of each segment is roughly the same. Therefore, in this embodiment, the segment used to calculate the information density in the target text is specified, the amount of information of the specified segment is calculated, and then the information density of the target text is determined based on the ratio of the amount of information of the specified segment to the text length of the text content contained in the specified segment.

[0144] For example: after determining the text content and sentences contained in the target text to be processed, the first 10% of the text content is designated as the designated segment for calculating the information density of the target text. Then the text content and sentences contained in the first 10% of the segment are determined, and the information volume of the designated segment is calculated. If the information volume of the designated segment is calculated to be 100 units of information volume, and the text length of the text content contained in the designated segment is 1000, then the information density of the target text is 100 / 1000=0.1 (unit information volume / unit length).

[0145] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0146] In one embodiment, the calculating the amount of information of the specified segment based on the text content and sentences contained in the specified segment includes:

[0147] Obtaining a word vector for each word in the text content;

[0148] Based on the word vector of each word in the text content, the words contained in each sentence and a preset convex hull algorithm, respectively calculate the convex hull volume of each sentence;

[0149] The sum of the convex hull volumes of the sentences is calculated, and the information content of the target text is determined according to the sum of the convex hull volumes.

[0150] The specific implementation process of this embodiment is the same as the above-mentioned calculation of the information amount of the target text based on the text content and each sentence of the target text, so it will not be repeated here.

[0151] In one embodiment, the calculating the amount of information of the specified segment based on the text content and sentences contained in the specified segment includes:

[0152] Determining the text length of the text content and the number of sentences included in the target text according to the text content and the sentences;

[0153] The product of the text length and the number of sentences is calculated, and the information volume of the target text is determined according to the product of the text length and the number of sentences.

[0154] The specific implementation process of this embodiment is the same as the above-mentioned calculation of the information volume of the target text based on the product of the text length of the target text and the number of sentences, so it will not be repeated here.

[0155] It should be noted that, based on the method for calculating the information amount of a specified segment shown in this embodiment, the information density of the target text can be calculated by the following formula:

[0156]

[0157] Where D represents the information density of the target text; I represents the information volume of the target text; L represents the text length of the target text; ΔI represents the information volume of the specified segment; ΔL represents the text length of the specified segment; Δn represents the text length of the specified segment. Therefore Therefore

[0158] Through this formula, the information density of the target text can be directly calculated by simply determining the number of sentences in the specified segment and the text length of the specified segment, thereby improving the calculation speed of the information density of the target text.

[0159] The following describes the specific process of obtaining the user's information reading rate and estimating the user's reading time for the target text.

[0160] Step 330: Obtain the user's information reading rate.

[0161] Step 340: based on the information volume of the target text and the information reading rate, estimate the user's reading time for the target text.

[0162] In one embodiment, obtaining the user's information reading rate includes:

[0163] Acquire the amount of information of the user's historical reading text and the reading time of the user for the historical reading text;

[0164] Based on the information volume of the historical reading text and the reading time, the user's information volume reading rate is obtained.

[0165] In this embodiment, the user's information volume reading rate is determined from the user's historical reading texts. Specifically, the information volume of the user's historical reading texts and the user's reading time for the historical reading texts are determined, and then the user's information volume reading rate is determined based on this.

[0166] Specifically, the user's information reading rate can be calculated by the following formula:

[0167]

[0168] Among them, ∑I represents the amount of information in the user's historical reading texts; ∑T represents the user's reading time for the historical reading texts; and v represents the user's information reading rate.

[0169] Furthermore, the reading time of the target text can be calculated by the following formula:

[0170]

[0171] Where p represents the target text p; t p represents the reading time of the user reading the target text p; I p represents the amount of information in the target text p; v represents the user's information reading rate.

[0172] Furthermore, by combining the information content of the target text determined based on the product of the text length and the number of sentences and related formulas, the following formula for calculating the reading time of the target text can be obtained:

[0173]

[0174] Among them, D p represents the information density of the target text p; L p represents the text length of the target text p; Δn p represents the number of sentences in the specified segment of the target text p; ΔL p Represents the text length of the specified segment in the target text p.

[0175] Through this formula, we can know that: based on the text length of the specified segment of the target text, the number of sentences in the specified segment of the target text, the text length of the target text, and the user's information reading rate, we can directly estimate the user's reading time for the target text.

[0176] In one embodiment, the user's historical reading records of texts are stored in the server.

[0177] In this example, the server can extract the user's historical reading text and send it to the execution subject of this embodiment, so that the execution subject calculates the amount of information in the user's historical reading text and the user's reading time of the historical reading text, and then obtains the user's information reading rate on this basis. The execution subject can be either a server or a user terminal.

[0178] In this embodiment, the server may calculate the amount of information in the user's historical reading text and the time the user read the historical reading text, and then send the amount of information in the historical reading text and the time the user read the historical reading text to the execution subject of this embodiment, and then obtain the user's information reading rate on this basis. The execution subject may be either a server or a user terminal.

[0179] The advantage of this embodiment is that the server centrally stores the records of users' historical reading texts, so that the server can centrally process the records of historical reading texts of batches of users, and then can perform further big data analysis on this basis to achieve more accurate reading time estimation.

[0180] In one embodiment, the user's historical reading text records are stored on the user side.

[0181] In this embodiment, the user terminal can extract the user's historical reading text and send it to the execution subject of this embodiment, so that the execution subject calculates the information volume of the user's historical reading text and the user's reading time of the historical reading text, and then obtains the user's information volume reading rate on this basis. The execution subject can be either a server or a user terminal.

[0182] In this embodiment, the user terminal may calculate the amount of information in the user's historical reading text and the time the user read the historical reading text, and then send the amount of information in the historical reading text and the time the user read the historical reading text to the execution subject of this embodiment, and then obtain the user's information reading rate on this basis. The execution subject may be either a server or a user terminal.

[0183] The advantage of this embodiment is that each user terminal stores the record of the historical reading text of the corresponding user, thereby improving the privacy security of the user.

[0184] In one embodiment, the client cannot connect to the server, that is, the client is offline. The execution subject of this embodiment can only be the client, and the text reading time can only be estimated by the client.

[0185] For example, a user wants to read an e-book stored offline locally on an offline client. The client determines the text content and sentences contained in the e-book that the user is trying to read, and then calculates the amount of information in the e-book on this basis. The client determines the user's information reading rate based on the records of the e-books that the user has read in the past stored in the client, and then estimates the user's reading time for the e-book.

[0186] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0187] In one embodiment, the user terminal can be connected to the server, and the execution subject of this embodiment is the server, but the record of the user's historical reading text is stored on the user terminal. In this embodiment, while the server calculates the information volume of the target text, the user terminal can be responsible for calculating the information volume of the user's historical reading text and the user's reading time for the historical reading text. The user terminal then uploads the user's historical reading text information volume and the user's reading time for the historical reading text to the server. Based on this, the server determines the user's information volume reading rate, and then estimates the user's reading time for the target text in combination with the information volume of the target text.

[0188] For example, if a user wants to read a paper pushed by a subscription account on the user side, the user's historical reading records are stored on the user side. Specifically, only the records of the user's historical reading of e-books are stored on the user side. While the server calculates the amount of information of the paper, the user side calculates the amount of information of these historically read e-books and the user's reading time of these historically read e-books based on the records of the user's historical reading of e-books stored on the user side, and sends the amount of information of these historically read e-books and the user's reading time of these historically read e-books to the server. On this basis, the server determines the user's information reading rate, and then estimates the user's reading time of the paper in combination with the amount of information in the paper.

[0189] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0190] In one embodiment, the user's historical reading text is the user's historical reading text within a predetermined period.

[0191] In this embodiment, the time point triggering the estimated target text reading time is used as the reference time point, and the user's historical reading texts within a predetermined period before the reference time point are obtained, and then the user's information volume reading rate is determined from the historical reading texts.

[0192] For example, the scheduled period is one week. The estimated target text reading time is triggered at "14:00 on August 19th", and the user's historical reading texts between "14:00 on August 12th" and "14:00 on August 19th" are obtained using this time point as the reference time point. The user's information volume reading rate is then determined from these historical reading texts.

[0193] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0194] In one embodiment, the historical reading texts of the user are all historical reading texts of the user.

[0195] In this embodiment, the time point triggering the estimated target text reading time is used as the reference time point, and all the user's historical reading texts before the reference time point are obtained, and then the user's information volume reading rate is determined from the historical reading texts.

[0196] For example, the estimated reading time of the target text is triggered at "14:00 on August 19th", and this time point is used as the reference time point to obtain all the historical reading texts of the user before "14:00 on August 19th". Then, the user's information reading rate is determined from these historical reading texts.

[0197] The advantage of this embodiment is that, as time goes by, more and more records of the user's historical reading of texts are accumulated, and accordingly, the estimation of the time taken by the user to read the target text will become more and more accurate.

[0198] Furthermore, in one embodiment, each time the estimated reading time of the target text is triggered, all the user's historical reading texts before the reference time point are obtained, and after the information volume of the historical reading texts and the user's reading time of the historical reading texts are determined, the reference time point of this time, the information volume of the historical reading texts and the user's reading time of the historical reading texts are recorded. When the estimated reading time of the target text is triggered next time, the user's historical reading texts between the reference time point of this time and the next reference time point are obtained, the information volume of the historical reading texts between the two reference time points and the corresponding reading time are determined, and are added to the recorded information volume and the corresponding reading time, respectively, so as to determine the information volume and the corresponding reading time of all the user's historical reading texts before the next reference time point.

[0199] For example, when the estimated target text reading time is triggered at "00:00 on 1 / 1", the information volume of all historical reading texts before "00:00 on 1 / 1" is determined to be 1000 units of information volume, and the corresponding reading time is 10 minutes, and the records are stored; the next time the estimated target text reading time is triggered at "00:00 on 4 / 5", all historical reading texts of the user between "00:00 on 1 / 1" and "00:00 on 4 / 5" are obtained, and the information volume of the historical reading texts during this period is determined to be 200 units of information volume, and the corresponding reading time is 2 minutes. According to the stored records, the information volume of all historical reading texts before "00:00 on 4 / 5" can be determined to be 1000+200=1200 units of information volume, and the corresponding reading time is 10+2=12 minutes, and the records are stored.

[0200] The advantage of this embodiment is that it avoids repeated calculation of already calculated parameters, reduces the amount of calculation, and improves calculation efficiency.

[0201] In one embodiment, obtaining the information reading rate of the user includes: obtaining the information reading rate of each user.

[0202] Based on the information volume of the target text and the information reading rate, estimating the reading time of the target text by the user, including: based on the information volume of the target text and the information reading rate of each user, estimating the reading time of each user for the target text.

[0203] In this embodiment, the information volume reading rate of each user terminal is determined for different users. Specifically, for a specific user, the information volume of the historical reading text of the specific user and the reading time of the historical reading text of the specific user are obtained, and then the information volume reading rate of the specific user is determined. Therefore, for the same target text, based on the information volume of the target text and the information volume reading time of each user, the reading time of each user for the target text is estimated respectively.

[0204] For example, a subscription account pushes the same target text to 100 subscribers of the subscription account. For each of the 100 subscribers, the amount of information of the subscriber's historical reading text and the subscriber's information reading rate are obtained, and then combined with the amount of information of the target text, the reading time of the subscriber for the target text is estimated.

[0205] The advantage of this embodiment is that it can provide personalized reading time estimation service for each user.

[0206] In one embodiment, obtaining the information reading rate of a user includes: obtaining the information reading rate of a target user cluster.

[0207] Based on the information volume of the target text and the information reading rate, estimating the user's reading time for the target text, including: based on the information volume of the target text and the information reading rate of the target user cluster, estimating the reading time of the target text for each user in the target user cluster.

[0208] The target user cluster refers to the user cluster where the users whose reading time is to be estimated belong.

[0209] In this embodiment, users are divided in advance to obtain various user clusters, and the information reading rate determined for any user in the same user cluster is the same. Specifically, the information reading rate of the target user cluster is the average of the information reading rates of each user in the target user cluster, representing the information reading rate of the user whose reading time is to be estimated.

[0210] In one embodiment, obtaining the information reading rate of the target user cluster includes:

[0211] Acquire the information volume of the historical reading text of at least one user in the target user cluster and the reading time of the historical reading text by the at least one user;

[0212] Calculating the sum of the information amounts of the historical reading texts corresponding to the target user cluster according to the information amount of the historical reading texts of the at least one user, and calculating the sum of the reading time corresponding to the user cluster according to the reading time of the historical reading texts by the at least one user;

[0213] The information volume reading rate corresponding to the target user cluster is obtained according to the sum of the information volume of the historical reading texts corresponding to the user cluster and the sum of the reading time corresponding to the user cluster.

[0214] In this embodiment, at least one user is selected from the target user cluster, and the amount of information of the historical reading texts and the corresponding reading time of the at least one user are obtained. Then, on this basis, the sum of the amount of information of the historical reading texts corresponding to the target user cluster and the sum of the corresponding reading time are determined, thereby determining the information volume reading rate corresponding to the target user cluster.

[0215] For example: a subscription account pushes the same target text to the target user cluster of the subscription account, and there are 100 subscribers in the target user cluster. 20 subscribers are randomly selected from the target user cluster, and the information volume and corresponding reading time of the historical reading texts of these 20 subscribers are obtained respectively, and then the sum of the information volume of the historical reading texts of these 20 subscribers and the sum of the corresponding reading time are determined. If the sum of the information volume of the historical reading texts of these 20 subscribers is 3000 units of information volume and the sum of the corresponding reading time is 300 units of time, then the corresponding information volume reading rate of the target user cluster is 3000 / 300=10 (unit information volume / unit time). Then the information volume reading rate of each subscriber in the target user cluster is 10 (unit information volume / unit time).

[0216] The advantage of this embodiment is that the reading time of the target user cluster is used as the reading time of each user in the target user cluster, which is more universal.

[0217] In one embodiment, the at least one user in the target user cluster is all users in the target user cluster.

[0218] It should be noted that this embodiment is only an exemplary description and should not limit the function and scope of use of the present disclosure.

[0219] In one embodiment, obtaining the user's information reading rate includes:

[0220] Determining the text category to which the target text belongs;

[0221] The information reading rate of the user for the text category is obtained.

[0222] It is understandable that due to the differences in the frequency of users' daily contact with various fields, the same user has different speeds of extracting and absorbing different types of information. Even if the amount of information in different types of text is the same, the same user has different reading rates for these texts. For example: the user is a physicist and has the most frequent contact with the field of physics. A text about gravity has an information volume of 1,000 units; a text about plant grafting also has an information volume of 1,000 units. But obviously, users read the former at a higher rate than the latter and spend less time.

[0223] Similarly, as different users are exposed to different fields on a daily basis, even with the same amount of information, different users have different speeds of extracting and absorbing different types of information. This results in different reading rates for texts with the same amount of information. For example: User A is a physicist, User B is a chemist, and the target text to be processed is a text about gravity. Due to the different fields they are exposed to on a daily basis, User A can read the text at a higher rate and in less time than User B.

[0224] Therefore, by specifically determining the user's reading rate of the information content of the text category to which the target text belongs, it is possible to more accurately estimate the user's reading time for the target text.

[0225] In one embodiment, obtaining the information reading rate of the user for the text category includes:

[0226] Acquire a target historical reading text belonging to the text category and the reading time of the target historical reading text by the user from the historical reading text of the user;

[0227] Based on the information volume of the target historical reading text and the reading time of the target historical reading text by the user, the information volume reading rate of the user for the text category is acquired.

[0228] The target historical reading text refers to a historical reading text used to determine the information volume reading rate of the user. Specifically, in this embodiment, the target historical reading text refers to a historical reading text that belongs to the same text category as the target text.

[0229] In this embodiment, after determining the text category to which the target text belongs, the target historical reading text of the text category is obtained from the user's historical reading texts, and then based on the amount of information in the target historical reading text and the user's reading time for the target historical reading text, the user's information reading rate for the text category is determined.

[0230] According to an embodiment of the present disclosure, Figure 7As shown, a device for estimating text reading time is also provided, the device comprising:

[0231] A determination module 410 is used to determine the text content and each sentence included in the target text to be processed;

[0232] An information volume calculation module 420, configured to calculate the information volume of the target text based on the text content and each of the sentences;

[0233] An acquisition module 430 is used to acquire a user's information reading rate;

[0234] The estimation module 440 is used to estimate the reading time of the target text by the user based on the information volume of the target text and the information reading rate.

[0235] In an exemplary embodiment of the present disclosure, the information volume calculation module 420 is configured as follows:

[0236] Obtaining a word vector for each word in the text content;

[0237] Based on the word vector of each word in the text content, the words contained in each sentence and a preset convex hull algorithm, respectively calculate the convex hull volume of each sentence;

[0238] The sum of the convex hull volumes of the sentences is calculated, and the information content of the target text is determined according to the sum of the convex hull volumes.

[0239] In an exemplary embodiment of the present disclosure, the information volume calculation module 420 is configured as follows:

[0240] Performing word segmentation on the text content to obtain a word segmentation result of the text content;

[0241] The word segmentation result is processed based on a pre-trained word vector model to obtain a word vector for each word in the text content.

[0242] In an exemplary embodiment of the present disclosure, the information volume calculation module 420 is configured as follows:

[0243] Determining the text length of the text content and the number of sentences included in the target text according to the text content and the sentences;

[0244] The product of the text length and the number of sentences is calculated, and the information volume of the target text is determined according to the product of the text length and the number of sentences.

[0245] In an exemplary embodiment of the present disclosure, the information volume calculation module 420 is configured as follows:

[0246] Based on the text content and the individual sentences, determining the information density of the target text and the text length of the text content;

[0247] The product of the information density and the text length is calculated, and the information volume of the target text is determined according to the product of the information density and the text length.

[0248] In an exemplary embodiment of the present disclosure, the information volume calculation module 420 is configured as follows:

[0249] Determining the text content and sentences included in the specified segment of the target text according to the text content and the sentences;

[0250] Calculate the information content of the specified segment based on the text content and sentences contained in the specified segment;

[0251] The ratio between the information amount of the designated segment and the text length of the text content contained in the designated segment is calculated, and the information density of the target text is determined according to the ratio.

[0252] In an exemplary embodiment of the present disclosure, the acquisition module 430 is configured as follows:

[0253] Acquire the amount of information of the user's historical reading text and the reading time of the user for the historical reading text;

[0254] Based on the information volume of the historical reading text and the reading time, the user's information volume reading rate is obtained.

[0255] In an exemplary embodiment of the present disclosure, the acquisition module 430 is configured to: acquire the information volume reading rate of each user;

[0256] The estimation module 440 is configured to estimate the reading time of each user for the target text based on the information volume of the target text and the information volume reading rate of each user.

[0257] In an exemplary embodiment of the present disclosure, the acquisition module 430 is configured to: acquire the information volume reading rate of the target user cluster;

[0258] The estimation module 440 is configured to estimate the reading time of the target text by each user in the target user cluster based on the information volume of the target text and the information volume reading rate of the target user cluster.

[0259] In an exemplary embodiment of the present disclosure, the acquisition module 430 is configured as follows:

[0260] Acquire the information volume of the historical reading text of at least one user in the target user cluster and the reading time of the historical reading text by the at least one user;

[0261] Calculating the sum of the information amounts of the historical reading texts corresponding to the user cluster according to the information amount of the historical reading texts of the at least one user, and calculating the sum of the reading time corresponding to the user cluster according to the reading time of the historical reading texts by the at least one user;

[0262] The information volume reading rate corresponding to the target user cluster is obtained according to the sum of the information volume of the historical reading texts corresponding to the user cluster and the sum of the reading time corresponding to the user cluster.

[0263] In an exemplary embodiment of the present disclosure, the acquisition module 430 is configured as follows:

[0264] Determining the text category to which the target text belongs;

[0265] The information reading rate of the user for the text category is obtained.

[0266] In an exemplary embodiment of the present disclosure, the acquisition module 430 is configured as follows:

[0267] Acquire a target historical reading text belonging to the text category and the reading time of the target historical reading text by the user from the historical reading text of the user;

[0268] Based on the information volume of the target historical reading text and the reading time of the target historical reading text by the user, the information volume reading rate of the user for the text category is acquired.

[0269] The method for estimating text reading time in the embodiment of the present disclosure can be implemented by the execution subject in the above description, which can be either a server or a user end. Figure 8 The execution subject 50 according to the embodiment of the present disclosure is described. Figure 8 The execution subject 50 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0270] like Figure 8 As shown, the execution subject 50 is in the form of a general-purpose computing device. The components of the execution subject 50 may include but are not limited to: at least one processing unit 510, at least one storage unit 520, and a bus 530 connecting different system components (including the storage unit 520 and the processing unit 510).

[0271] The storage unit stores program codes, which can be executed by the processing unit 510, so that the processing unit 510 performs the steps according to various exemplary embodiments of the present invention described in the description of the exemplary method described above in this specification. For example, the processing unit 510 can perform the following steps: Figure 3 The steps shown in .

[0272] The storage unit 520 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 5201 and / or a cache storage unit 5202 , and may further include a read-only storage unit (ROM) 5203 .

[0273] The storage unit 520 may also include a program / utility 5204 having a set (at least one) of program modules 5205, such program modules 5205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0274] Bus 530 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0275] The executive body 50 may also communicate with one or more external devices 600 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the executive body 50, and / or communicate with any device that enables the executive body 50 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 550. In addition, the executive body 50 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 560. As shown, the network adapter 560 communicates with other modules of the executive body 50 via a bus 530. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the executive body 50, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0276] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0277] In an exemplary embodiment of the present disclosure, a computer program medium is further provided, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor of a computer, the computer is enabled to execute the method described in the above method embodiment.

[0278] According to one embodiment of the present disclosure, a program product for implementing the method in the above method embodiment is also provided, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto, and in this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0279] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RGM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0280] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0281] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0282] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as JScript, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0283] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0284] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0285] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0286] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A method for estimating text reading time, characterized in that: The method comprises: Determine the text content and each sentence contained in the target text to be processed; Calculating the information content of the target text based on the text content and the various sentences; Get the user's information reading rate; estimating the user's reading time for the target text based on the information volume of the target text and the information reading rate; The step of determining the text content and each sentence contained in the target text to be processed includes: Determining the text content contained in the target text; The text content is divided into sentences to obtain sentence division results of the text content, and the contents of each sentence included in the target text are determined from the sentence division results.

2. The method according to claim 1, characterized in that The calculating the information amount of the target text based on the text content and the various sentences includes: Obtaining a word vector for each word in the text content; Based on the word vector of each word in the text content, the words contained in each sentence and a preset convex hull algorithm, respectively calculate the convex hull volume of each sentence; The sum of the convex hull volumes of the sentences is calculated, and the information content of the target text is determined according to the sum of the convex hull volumes.

3. The method according to claim 2, characterized in that The obtaining of a word vector for each word in the text content includes: Performing word segmentation on the text content to obtain a word segmentation result of the text content; The word segmentation result is processed based on a pre-trained word vector model to obtain a word vector for each word in the text content.

4. The method according to claim 1, characterized in that: The calculating the information amount of the target text based on the text content and the various sentences includes: Determining the text length of the text content and the number of sentences included in the target text according to the text content and the sentences; The product of the text length and the number of sentences is calculated, and the information volume of the target text is determined according to the product of the text length and the number of sentences.

5. The method according to claim 1, characterized in that The calculating the information amount of the target text based on the text content and the various sentences includes: Based on the text content and the individual sentences, determining the information density of the target text and the text length of the text content; The product of the information density and the text length is calculated, and the information volume of the target text is determined according to the product of the information density and the text length.

6. The method according to claim 5, characterized in that The determining the information density of the target text based on the text content and the individual sentences includes: Determining the text content and sentences included in the specified segment of the target text according to the text content and the sentences; Calculate the information content of the specified segment based on the text content and sentences contained in the specified segment; The ratio between the information amount of the designated segment and the text length of the text content contained in the designated segment is calculated, and the information density of the target text is determined according to the ratio.

7. The method according to claim 1, characterized in that The obtaining of the user's information reading rate includes: Acquire the amount of information of the user's historical reading text and the reading time of the user for the historical reading text; Based on the information volume of the historical reading text and the reading time, the user's information volume reading rate is obtained.

8. The method according to claim 1, characterized in that The obtaining of the information reading rate of the user includes: obtaining the information reading rate of each user; The estimating the reading time of the target text by the user based on the information volume of the target text and the information reading rate includes: estimating the reading time of the target text by each user based on the information volume of the target text and the information reading rate of each user.

9. The method according to claim 1, characterized in that: The obtaining of the information reading rate of the user includes: obtaining the information reading rate of the target user cluster; The estimating the reading time of the target text by the user based on the information volume of the target text and the information reading rate includes: estimating the reading time of the target text by each user in the target user cluster based on the information volume of the target text and the information reading rate of the target user cluster.

10. The method according to claim 9, characterized in that The step of obtaining the information reading rate of the target user cluster includes: Acquire the information volume of the historical reading text of at least one user in the target user cluster and the reading time of the historical reading text by the at least one user; Calculating the sum of the information amounts of the historical reading texts corresponding to the user cluster according to the information amount of the historical reading texts of the at least one user, and calculating the sum of the reading time corresponding to the user cluster according to the reading time of the historical reading texts by the at least one user; The information volume reading rate corresponding to the target user cluster is obtained according to the sum of the information volume of the historical reading texts corresponding to the user cluster and the sum of the reading time corresponding to the user cluster.

11. The method according to any one of claims 1 to 10, characterized in that The obtaining of the user's information reading rate includes: Determining the text category to which the target text belongs; The information reading rate of the user for the text category is obtained.

12. The method according to claim 11, characterized in that The obtaining of the user's information reading rate for the text category includes: Acquire a target historical reading text belonging to the text category and the reading time of the target historical reading text by the user from the historical reading text of the user; Based on the information volume of the target historical reading text and the reading time of the target historical reading text by the user, the information volume reading rate of the user for the text category is acquired.

13. A device for estimating text reading time, characterized in that: The device comprises: A determination module, used to determine the text content and each sentence contained in the target text to be processed; An information volume calculation module, used for calculating the information volume of the target text based on the text content and each sentence; An acquisition module is used to acquire the user's information reading rate; An estimation module, used for estimating the reading time of the target text by the user based on the information volume of the target text and the reading speed of the information volume; The step of determining the text content and each sentence contained in the target text to be processed includes: Determining the text content contained in the target text; The text content is divided into sentences to obtain sentence division results of the text content, and the contents of each sentence included in the target text are determined from the sentence division results.

14. An electronic device for estimating text reading time, characterized in that: include: a memory storing computer-readable instructions; The processor reads the computer readable instructions stored in the memory to execute the method according to any one of claims 1 to 12.

15. A computer program medium, characterized in that Computer readable instructions are stored thereon, and when the computer readable instructions are executed by a processor of a computer, the computer is caused to execute the method of any one of claims 1-12.

Images