Resume content extraction method and device

Abstract

The invention provides a resume content extraction method and device, and the method comprises the steps: obtaining a target resume, and extracting at least one text box in the target resume; based on the at least one textbox, identifying a column type of the target resume; under the condition that the column type is multiple columns, determining a target reading sequence of each textbox according to a preset inter-column reading sequence and an in-column reading sequence; and based on the target reading sequence, reading the text content from each textbox in sequence to obtain the resume content of the target resume. According to the method, the recognition rate of the column type can be effectively improved, the chaos caused by directly extracting the resume content is avoided, the sequence and efficiency of extracting the resume content are improved, and the accuracy of the resume content is also improved.

Description

[0001] This application claims the priority of the Chinese patent application filed with the China Patent Office on August 11, 2023, with application number "202311017073.7" and invention name "Resume Content Extraction Method and Device", all contents of which are recorded in this application by reference. Technical Field

[0002] The present application relates to the field of text processing technology, and in particular to a resume content extraction method, a resume content extraction device, a computing device, and a computer-readable storage medium. Background Art

[0003] In the information age, the amount of information on the Internet has increased dramatically, the total amount of information has increased, and the ways to obtain information have increased, but invalid information has also increased significantly, resulting in a decrease in people's reading efficiency. How to obtain valuable information from massive amounts of information in the presence of information interference is particularly important.

[0004] Resumes are a very common and important text. Information extraction from resumes must ensure both the accuracy of text information extraction and the order of text information. In the prior art, information in resumes is usually extracted directly. However, this method may result in incomplete or chaotic information extraction from resumes due to the variety of resume formats. Therefore, an effective solution is urgently needed to solve the above problems. Summary of the invention

[0005] In view of this, the embodiment of the present application provides a resume content extraction method to solve the technical defects existing in the prior art. The embodiment of the present application also provides a resume content extraction device, a computing device, and a computer-readable storage medium.

[0006] According to a first aspect of an embodiment of the present application, a resume content extraction method is provided, comprising:

[0007] Obtain a target resume and extract at least one text box in the target resume;

[0008] Based on at least one text box, identifying the column type of the target resume;

[0009] When the column type is multiple columns, the target reading order of each text box is determined according to the preset inter-column reading order and intra-column reading order;

[0010] Based on the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume.

[0011] According to a second aspect of an embodiment of the present application, a resume content extraction device is provided, comprising:

[0012] An acquisition module is configured to acquire a target resume and extract at least one text box in the target resume;

[0013] A recognition module configured to recognize a column type of a target resume based on at least one text box;

[0014] The first determination module is configured to determine the target reading order of each text box according to a preset inter-column reading order and an intra-column reading order when the column type is multiple columns;

[0015] The reading module is configured to read the text content from each text box in turn based on the target reading order to obtain the resume content of the target resume.

[0016] According to a third aspect of an embodiment of the present application, a computing device is provided, including:

[0017] Memory and processor;

[0018] The memory is used to store computer-executable instructions, and the processor implements the steps of the resume content extraction method when executing the computer-executable instructions.

[0019] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the resume content extraction method are implemented.

[0020] According to a fifth aspect of an embodiment of the present application, a chip is provided, which stores computer instructions, and when the computer instructions are executed by the chip, the steps of the resume content extraction method are implemented.

[0021] The resume content extraction method provided by the present application extracts at least one text box in the target resume; then identifies the column type of the target resume based on the at least one text box; and when the column type is multiple columns, determines the target reading order of each text box according to the preset inter-column reading order and intra-column reading order; and then reads the text content from each text box in turn based on the target reading order to obtain the resume content of the target resume. Using at least one text box to identify the column type of the target resume can effectively improve the recognition rate of the column type; when the column type is multiple columns, the target reading order of each text box is determined based on the inter-column reading order and the intra-column reading order, avoiding the confusion of directly extracting the resume content, which not only improves the orderliness and efficiency of extracting the resume content, but also improves the accuracy of the resume content. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a structural diagram of a resume content extraction system provided by an embodiment of the present application;

[0023] Figure 2 is a flow chart of a resume content extraction method provided by an embodiment of the present application;

[0024] Figure 3A This is a schematic diagram of the effect of extracting a text box provided by an embodiment of the present application;

[0025] Figure 3B is a schematic diagram of a target resume provided by an embodiment of the present application;

[0026] Figure 3C is a schematic diagram of another target resume provided by an embodiment of the present application;

[0027] Figure 3D is a flow chart of a first column type identification method provided by an embodiment of the present application;

[0028] Figure 3E This is a schematic diagram of a projection processing effect provided by an embodiment of the present application;

[0029] Figure 4 is a flow chart of a second column type identification method provided in an embodiment of the present application;

[0030] Figure 5 is a flow chart of a third column type identification method provided in an embodiment of the present application;

[0031] Figure 6 is a flow chart of a single column determination method provided by an embodiment of the present application;

[0032] Figure 7 is a flow chart of a method for determining a column reading order provided by an embodiment of the present application;

[0033] Figure 8 is a flow chart of another method for determining a target reading order provided by an embodiment of the present application;

[0034] Fig. 9 is a flowchart of a text content reading method provided by an embodiment of the present application;

[0035] Fig.10 is a flow chart of a lateral calibration method provided by an embodiment of the present application;

[0036] Fig.11 is a flow chart of another lateral calibration method provided by an embodiment of the present application;

[0037] Fig.12 This is a flow chart of a method for obtaining a target resume provided by an embodiment of the present application;

[0038] Fig.13It is a processing flow chart of a resume content extraction method provided in one embodiment of the present application;

[0039] Fig.14 It is a structural schematic diagram of a resume content extraction device provided in one embodiment of the present application;

[0040] Fig.15 It is a structural block diagram of a computing device 1500 provided in one embodiment of the present application. DETAILED DESCRIPTION

[0041] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0042] The terms used in one or more embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present application. The singular forms of "a", "said" and "the" used in one or more embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present application refers to any or all possible combinations of one or more associated listed items.

[0043] It should be understood that, although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present application, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first.

[0044] First, the terms involved in one or more embodiments of the present invention are explained.

[0045] Optical Character Recognition (OCR) refers to the process in which an electronic device (such as a scanner or digital camera) examines characters printed on paper, determines their shape by detecting dark and light patterns, and then uses character recognition methods to translate the shape into computer text.

[0046] Portable Document Format (PDF) is an electronic file format based on a language image model.

[0047] PDFMiner: PDFMiner is a tool for extracting information from PDF documents.

[0048] In the present application, a resume content extraction method is provided. The present application also relates to a resume content extraction device, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0049] Figure 1 It is a structural diagram of a resume content extraction system provided according to an embodiment of the present application.

[0050] The execution subject of the resume content extraction method provided in the embodiment of the present application can be a server or a terminal, which is not limited in the embodiment of the present application. In addition, the terminal can be any electronic product that can interact with the user, such as a PC (Personal Computer), a mobile phone, a PPC (Pocket PC), a tablet computer, etc. The server can be a single server, a server cluster composed of multiple servers, or a cloud computing service center, which is not limited in the embodiment of the present application.

[0051] Taking the execution subject as a terminal as an example, the terminal obtains the target resume from the server and extracts at least one text box in the target resume; then, based on the at least one text box, the column type of the target resume is identified; then, if the column type is multiple columns, the target reading order of each text box is determined according to the preset inter-column reading order and intra-column reading order; based on the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume. In addition, the terminal can also obtain the target resume locally from the terminal.

[0052] Taking the execution subject as a server as an example, after receiving the target resume uploaded by the user through the terminal, the server extracts at least one text box in the target resume; then identifies the column type of the target resume based on the at least one text box; then, if the column type is multi-column, determines the target reading order of each text box according to the preset inter-column reading order and intra-column reading order; based on the target reading order, reads the text content from each text box in turn to obtain the resume content of the target resume. In addition, the server can also obtain the target resume locally from the server.

[0053] In an embodiment of the present application, at least one text box in a target resume is extracted; then based on the at least one text box, the column type of the target resume is identified; and when the column type is multiple columns, the target reading order of each text box is determined according to a preset inter-column reading order and intra-column reading order; and then based on the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume. Using at least one text box to identify the column type of the target resume can effectively improve the recognition rate of the column type; when the column type is multiple columns, the target reading order of each text box is determined based on the inter-column reading order and the intra-column reading order, avoiding the confusion of directly extracting the resume content, which not only improves the orderliness and efficiency of extracting the resume content, but also improves the accuracy of the resume content.

[0054] Figure 2 A flowchart of a resume content extraction method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0055] Step 202: Obtain a target resume and extract at least one text box in the target resume.

[0056] The execution subject of the resume content extraction method can be a computing device with resume content extraction function, such as a server, terminal, etc. with resume content extraction function.

[0057] Specifically, a resume refers to a resume document edited by an applicant, which may include basic information, job search intention, work experience, education experience, research direction, academic achievements, self-evaluation, etc. A target resume refers to a resume whose content is to be extracted.

[0058] In actual applications, the target resume can be a resume in any text format, such as a document format (DOC, Document), a text format (txt), an image format, or a portable document format (PDF, Portable DocumentFormat), which is not limited in this application. There are many ways to obtain the target resume. In specific implementation, a user may send an acquisition instruction of the target resume to the execution subject, and accordingly, the execution subject starts to acquire the target resume after receiving the acquisition instruction; or the execution subject may automatically acquire the target resume of the resume content to be extracted at preset time intervals. For example, after a preset time, a server with a resume content extraction function automatically acquires the target resume in a designated access area; or after a preset time, a terminal with a resume content extraction function automatically acquires the target resume stored locally. This specification does not impose any limitations on the method of acquiring the target resume.

[0059] Further, based on obtaining the target resume whose resume content is to be extracted, the execution subject performs an extraction operation on the target resume, thereby extracting at least one text box containing text content in the target resume.

[0060] In a possible implementation of the embodiments of the present specification, after obtaining the target resume, a corresponding text box extraction tool can be selected according to the text format of the target resume, and then the text box extraction tool can be used to extract the text box from the target resume, where the text box contains the text content that constitutes the resume content. In this way, by selecting the text box extraction tool corresponding to the text format of the target resume and extracting the text box, the accuracy and speed of extracting the text box can be improved. Figure 3A , Figure 3A A schematic diagram of the effect of extracting a text box provided by an embodiment of the present application is shown, wherein each black rectangle represents a text box.

[0061] For example, if the target resume obtained is in PDF format, then the pdfominer tool corresponding to the PDF format is selected to perform an extraction operation on the target resume, thereby extracting at least one text box containing text content in the target resume.

[0062] For another example, if the target resume is in an image format, an optical character recognition tool (OCR) corresponding to the image format is selected to perform an extraction operation on the target resume, thereby extracting at least one text box containing text content in the target resume.

[0063] Step 204: Based on at least one text box, identify the column type of the target resume.

[0064] Specifically, the column type refers to the type of layout of the resume content file, which is divided into single column, double column, triple column, etc., that is, single column and multiple columns. Single column, that is, one column, the input content is filled from left to right or from right to left, and it automatically returns when it reaches the right boundary or left boundary. Multiple columns, the input content is filled from left to right or from right to left, and it automatically returns when the column distance is reached; after the current column is full, it automatically enters the next column. It should be noted that for any line of any column, the line can have one text box or multiple text boxes, and this application does not limit this.

[0065] In a possible implementation of the embodiment of the present specification, after extracting all text boxes of the target resume, the overall arrangement and layout of all text boxes can be determined first, and then the column type of the target resume can be identified according to the overall arrangement and layout, that is, whether the target resume is single-column or multi-column. In this way, the efficiency of identifying the column type can be improved.

[0066] For example, based on the overall arrangement layout of all text boxes in the target resume, including the position of the text boxes, the arrangement order between the text boxes, and the distance between the text boxes, it is identified whether the target resume is single-column or multi-column.

[0067] In a possible implementation of the embodiment of the present specification, after extracting all text boxes of the target resume, a two-dimensional coordinate system can be established, and then the coordinates of each text box can be determined, and then the column type of the target resume can be identified according to the coordinates of each text box, that is, whether the target resume has a single column or multiple columns. In this way, the accuracy of identifying the column type can be improved.

[0068] Preferably, the coordinates of the text box are (a, b, w, h), where a represents the horizontal coordinate of the lower left corner of the current text box, that is, the coordinate on the horizontal axis, b represents the vertical coordinate of the lower left corner of the current text box, that is, the coordinate on the vertical axis, w represents the width of the current text box, and h represents the height of the current text box.

[0069] See also Figure 3B , Figure 3B A schematic diagram of a target resume provided by an embodiment of the present application is shown: the target resume includes five text boxes, among which the coordinates of text box A1 are (0, 0, 1, 1), the coordinates of text box A2 are (3, 0, 2, 0.5), the coordinates of text box A3 are (0, 2, 1, 1), the coordinates of text box A4 are (3, 2, 1, 1), and the coordinates of text box A5 are (0, 4, 0.5, 1). Therefore, it can be determined that text box A1, text box A3 and text box A5 are one column, and text box A2 and text box A4 are one column. Then the column type of the target resume is double column, that is, multiple columns.

[0070] Step 206: When the column type is multiple columns, determine the target reading order of each text box according to the preset inter-column reading order and intra-column reading order.

[0071] Specifically, the inter-column reading order refers to the reading order between columns, for example, from left to right, the first column is read first, and after the first column is read, the second column is read, and so on, until the last column is read; or the last column is read first, and after the last column is read, the second to last column is read, and so on, until the first column is read. The intra-column reading order refers to the reading order for all text boxes in each column, for example, from top to bottom, from the first row of text boxes to the last row of text boxes. It should be noted that when there are multiple text boxes in a row of a column, the reading order of the text boxes in the row can be from left to right. Preferably, the intra-column reading order is first from left to right and then from top to bottom. The target reading order refers to the final reading order of the text boxes determined by combining the inter-column reading order and the intra-column reading order.

[0072] In actual application, when it is recognized that the column type of the target resume is multi-column, the preset inter-column reading order and intra-column reading order are obtained, and then the target reading order of each text box in each column is determined according to the inter-column reading order and the intra-column reading order: first determine which column's text box to read first and which column's text box to read next according to the inter-column reading order; determine the reading order of each text box in the current column according to the intra-column reading order. That is, for the current column, read the text boxes in the current column according to the intra-column reading order, then read the next column, read the text boxes in the next column according to the intra-column reading order, and then continue to read the next column, and so on.

[0073] For example, the reading order between columns is from left to right, and the reading order within a column is from left to right and then from top to bottom. See 3C, Figure 3C This is a schematic diagram of another target resume provided by an embodiment of the present application: the target resume has two columns, a left column and a right column, wherein the left column has five text boxes: text box a11 (first row), text box a12 (first row), text box a21 (second row), text box a31 (third row) and text box a32 (third row), and the right column has three text boxes: text box b11 (first row), text box b21 (second row) and text box b31 (third row), then according to the reading order between columns and the reading order within a column, the target reading order of each text box can be determined as: text box a11, text box a12, text box a21, text box a31, text box a32, text box b11, text box b21, text box b31.

[0074] It should be noted that the inter-column reading order and the intra-column reading order can be set according to demand, and this application does not limit this.

[0075] When the column type is multiple columns, the target reading order of each text box is determined based on the reading order between columns and the reading order within a column, thus avoiding the confusion of directly extracting the resume content. This not only improves the orderliness and efficiency of extracting the resume content, but also improves the accuracy of the resume content.

[0076] Step 208: Based on the target reading order, read the text content from each text box in turn to obtain the resume content of the target resume.

[0077] Specifically, text content refers to the text contained in the text box; resume content refers to the content pointed to by the target resume, that is, the collection of text contents in each text box.

[0078] In actual applications, after the target reading order of each text box is determined, the text content in each text box is read in sequence according to the target reading order. That is, the text content read according to the target reading order is the resume content of the target resume.

[0079] Using the above example, the text content of each text box is shown in Table 1. According to the target reading order, the text content "name" is read from text box a11, the text content "Xiaohong" is read from text box a12, the text content "work experience" is read from text box a21, the text content "company name" is read from text box a31, the text content "Aaa" is read from text box a32, the text content "specialty" is read from text box b11, the text content "singing" is read from text box b21, and the text content "painting" is read from text box b31. Then the resume content of the target resume is "name Xiaohong work experience company name Aaa specialty singing painting".

[0080] Table 1 Text content of each text box

[0081]

[0082] In the above implementation, at least one text box in the target resume is extracted; then based on the at least one text box, the column type of the target resume is identified; and when the column type is multiple columns, the target reading order of each text box is determined according to the preset inter-column reading order and intra-column reading order; then based on the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume. Using at least one text box to identify the column type of the target resume can effectively improve the recognition rate of the column type; when the column type is multiple columns, the target reading order of each text box is determined based on the inter-column reading order and the intra-column reading order, avoiding the confusion of directly extracting the resume content, which not only improves the orderliness and efficiency of extracting the resume content, but also improves the accuracy of the resume content.

[0083] Figure 3D A flowchart of a first column type identification method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0084] Step 302: Establish a plane coordinate system in the target resume, wherein the X-axis of the plane coordinate system is the extension direction of the text lines in the target resume, and the Y-axis is the arrangement direction of the text lines in the target resume.

[0085] Step 304: Project at least one text box onto the X-axis to obtain at least one projection line segment of the at least one text box on the X-axis.

[0086] Step 306: Identify the column type of the target resume based on at least one projection line segment.

[0087] Specifically, a plane coordinate system is also a plane rectangular coordinate system, also known as a rectangular coordinate system, which is composed of two axes that are perpendicular to each other and have a common origin on the same plane - the X-axis and the Y-axis, wherein the X-axis is a horizontal axis, also known as the horizontal axis. In this application, since the extension direction of the text line in the target resume refers to the arrangement direction of the same line of text in a logical order, that is, the horizontal direction, such as horizontal to the right and horizontal to the left, the X-axis is the extension direction of the text line in the target resume; the Y-axis is a vertical axis, also known as the vertical axis. In this application, since the arrangement direction of the text line in the target resume refers to the arrangement direction of different lines of text in a logical order, that is, the vertical direction, the Y-axis is the arrangement direction of the text line in the target resume. Projection refers to the shadow of a figure projected onto a surface or a line, and in this application, the shadow of the text box is projected onto the X-axis. Projection line segments refer to the line segments formed by all text boxes projected onto the X-axis.

[0088] In actual application, there are many ways to establish a plane coordinate system in the target resume. For example, it can be a plane coordinate system established with the upper left corner of the target resume as the origin, the extension direction of the text line in the target resume as the X-axis, and the arrangement direction of the text line in the target resume as the Y-axis; it can be a plane coordinate system established with the lower left corner of the target resume as the origin, the extension direction of the text line in the target resume as the X-axis, and the arrangement direction of the text line in the target resume as the Y-axis; it can be a plane coordinate system established with any point on the left border of the target resume as the origin, the extension direction of the text line in the target resume as the X-axis, and the arrangement direction of the text line in the target resume as the Y-axis; it can also be a plane coordinate system established with any point on the target resume as the origin, the extension direction of the text line in the target resume as the X-axis, and the arrangement direction of the text line in the target resume as the Y-axis. This specification does not impose any restrictions on the method of establishing a plane coordinate system in the target resume.

[0089] Furthermore, on the basis of establishing the plane coordinate system, all text boxes are projected onto the X-axis to obtain the overall projection result of all text boxes on the X-axis, that is, at least one projection line segment. Then, the column type of the target resume can be identified according to the number of at least one projection line segment, and the column type of the target resume can also be identified according to the length of at least one projection line segment. This specification does not impose any limitation on the method of identifying the column type according to at least one projection line segment.

[0090] See also Figure 3E , Figure 3EA schematic diagram of the effect of a projection processing provided by an embodiment of the present application is shown: preferably, the origin position is determined according to the leftmost text box and the bottommost text box in the target resume, that is, the intersection of the straight line where the left boundary of the leftmost text box is located and the target parallel line is taken as the origin, wherein the target parallel line and the straight line where the bottom boundary of the bottommost text box is located are parallel to each other and are located below the straight line, a plane coordinate system is established with the extension direction of the text line in the target resume as the X-axis and the arrangement direction of the text line in the target resume as the Y-axis, and then each text box is projected onto the X-axis, that is, projected from top to bottom, to obtain at least one projection line segment, such as Figure 3E In the figure above, three projection line segments are obtained, such as Figure 3E In the figure below, a projection line segment is obtained.

[0091] It should be noted that step 302 to step 306 are a specific implementation of step 204 above.

[0092] In an embodiment of the present application, a plane coordinate system is first established in the target resume, and then at least one text box is projected onto the X-axis to obtain at least one projection line segment of the at least one text box on the X-axis, and then the column type of the target resume is identified based on the at least one projection line segment. This not only enables each text box to be accurately located, but also allows the column type of the target resume to be identified by projecting all the text boxes together, thereby avoiding the data processing required to analyze and identify the column type one by one due to the number of text boxes. While reducing the data processing amount and improving the accuracy of the projection line segments, the accuracy and speed of identifying the column type are further improved, thereby improving the efficiency of extracting resume content.

[0093] Figure 4 A flow chart of a second column type identification method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0094] Step 402: When the number of at least one projection line segment is one, determine that the column type of the target resume is single column.

[0095] Step 404: When the number of at least one projection line segment is plural, determine that the column type of the target resume is multi-column.

[0096] In practical applications, it is necessary to identify the column type of the target resume based on the number of at least one projection line segment: when there is only one projection line segment, it means that there is no gap between the horizontal projections of the text boxes in the target resume, that is, the text boxes belong to the same column, and the column type of the target resume is single column; when there are multiple projection line segments, it means that there are gaps between the horizontal projections of the text boxes in the target resume, which means that the column type of the target resume can be multiple columns.

[0097] See also Figure 3E , Figure 3E FIG. 1 is a schematic diagram showing the effect of a projection processing provided by an embodiment of the present application. Figure 3E In the figure above, three projection line segments are obtained, indicating that the column type of the target resume corresponding to the figure above is three columns, that is, multiple columns; Figure 3E In the figure below, we get a projection line segment, which means that the column type of the target resume corresponding to the figure below is single column.

[0098] It should be noted that step 402 to step 404 are a specific implementation of step 306 above.

[0099] In an embodiment of the present application, the column type of the target resume is identified based on the number of projection line segments. The column type of a target resume with one item is determined as a single column, and the column type of a target resume with multiple items is determined as multiple columns. This is simple and easy to implement, and the column type can be determined only based on the number of projection line segments, which greatly improves the speed of determining the column type and further improves the efficiency of extracting resume content.

[0100] Figure 5 A flowchart of a third column type identification method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0101] Step 502: when there are multiple projection line segments, determine the total distance between the starting point of the first projection line segment and the ending point of the last projection line segment.

[0102] Step 504: Determine the interval distance between two adjacent projection line segments in at least one projection line segment, and respectively determine the ratio of each interval distance to the total distance.

[0103] Step 506: When there is a target ratio among the ratios, determine that the column type of the target resume is multi-column, wherein the target ratio is a ratio greater than a first threshold value.

[0104] Specifically, the first projection line refers to the leftmost projection line segment along the X-axis direction; the starting point refers to the left endpoint of the projection line segment on the X-axis. The last projection line segment refers to the rightmost projection line segment along the X-axis direction. The end point refers to the right endpoint of the projection line segment on the X-axis. The interval distance refers to the distance between two adjacent projection line segments, that is, the distance between the end point of the left projection line segment and the starting point of the right projection line segment in the two adjacent projection line segments, that is, the length of the zero value in the projection result. The first threshold is a pre-set value used to measure whether the interval distance represents a column division, and can be adjusted according to needs.

[0105] In practical applications, when there are multiple projection line segments, which text boxes are located in the same column and which text boxes can be located in different columns can be determined based on the interval distance between the projection line segments. Specifically, the coordinates of the starting point and the ending point of each projection line segment are determined on the X-axis; then the coordinates of the ending point of the last projection line segment are subtracted from the coordinates of the starting point of the first projection line segment to obtain the total distance; then, for every two adjacent projection line segments, the coordinates of the starting point of the right projection line segment are subtracted from the coordinates of the ending point of the left projection line segment to determine the interval distance and obtain at least one interval distance. Then, the ratio of each interval distance to the total distance is calculated respectively. When there is a ratio greater than the first threshold, the column type of the target resume is determined to be multiple columns.

[0106] See also Figure 3E From the above figure, three projection line segments are obtained. The coordinates of the starting point and the end point of each projection line segment are shown in Table 2. The total distance between the starting point of the first projection line segment and the end point of the third projection line segment is 15-10=15, the first interval distance between the starting point of the second projection line segment and the end point of the first projection line segment is 6-5.5=0.5, and the second interval distance between the starting point of the third projection line segment and the end point of the second projection line segment is 10-9=1. The first ratio is 0.5 / 15=1 / 30, and the second ratio is 1 / 15. Assuming that the first threshold is 1 / 20, 1 / 15 is greater than 1 / 20, then there is a target ratio, and the column type of the target resume is multi-column.

[0107] Table 2 Coordinates of the starting and ending points of each projection line segment

[0108] Projection line segment Starting point End point First projected line segment 0 5.5 Second projection line segment 6 9 The third projection line segment 10 15

[0109] In addition, the number of columns or columns of the target resume can also be determined according to the number of target ratios, that is, the number of columns of the target resume is the sum of the number of target ratios plus one. For example, if there is one target ratio, the number of columns of the target resume is two, that is, the target resume is a double column; if there are three target ratios, the number of columns of the target resume is four, that is, the target resume is a four column. In this way, the specific number of columns of the target resume can be accurately determined, which is conducive to improving the efficiency of determining the target reading order based on the inter-column reading order and the intra-column reading order, and further the efficiency of extracting the resume content.

[0110] It should be noted that step 502 to step 506 are a specific implementation of step 404 above.

[0111] In the embodiment of the present application, when there are multiple projection line segments, the ratio of each interval distance to the total distance is compared with the first threshold value, and when there is a target ratio greater than the first threshold value, the column type of the target resume is determined to be multi-column. In this way, the column type is determined in combination with the number of projection line segments and the first threshold value, which avoids the error of identifying the column type directly based on the number of projection line segments when there are slight intervals between the projections of the text boxes in the target resume on the X-axis but the column type is a single column, thereby improving the accuracy of determining the column type to a certain extent and further extracting the efficiency of the resume content.

[0112] Figure 6 A flowchart of a single column determination method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0113] Step 602: when there are multiple projection line segments, determine the total distance between the starting point of the first projection line segment and the ending point of the last projection line segment.

[0114] Step 604: Determine the interval distance between two adjacent projection line segments in at least one projection line segment, and respectively determine the ratio of each interval distance to the total distance.

[0115] Step 606: When the target ratio does not exist in the ratios, determine that the column type of the target resume is a single column.

[0116] Figure 6 Steps 602 and 604 are Figure 5 Steps 502 and 504 are the same and will not be repeated here.

[0117] Specifically, in the case of multiple projection line segments, the coordinates of the starting point and the ending point of each projection line segment are determined on the X-axis respectively; then the coordinates of the ending point of the last projection line segment are subtracted from the coordinates of the starting point of the first projection line segment to obtain the total distance; then, for each two adjacent projection line segments, the coordinates of the starting point of the right projection line segment are subtracted from the coordinates of the ending point of the left projection line segment to determine the interval distance, and at least one interval distance is obtained. Then, the ratio of each interval distance to the total distance is calculated respectively, and when there is no ratio greater than the first threshold, the column type of the target resume is determined to be a single column.

[0118] See also Figure 3EFrom the above figure, three projection line segments are obtained. The coordinates of the starting point and the end point of each projection line segment are shown in Table 2. The total distance between the starting point of the first projection line segment and the end point of the third projection line segment is 15-10=15, the first interval distance between the starting point of the second projection line segment and the end point of the first projection line segment is 6-5.5=0.5, and the second interval distance between the starting point of the third projection line segment and the end point of the second projection line segment is 10-9=1. The first ratio is 0.5 / 15=1 / 30, and the second ratio is 1 / 15. Assuming that the first threshold is 1 / 10, 1 / 15 is less than 1 / 10, and 1 / 30 is less than 1 / 10, that is, there is no target ratio, indicating that the column type of the target resume is single column.

[0119] It should be noted that step 602 to step 606 are another specific implementation of step 404 above.

[0120] In the embodiment of the present application, when there are multiple projected line segments, the ratio of each interval distance to the total distance is compared with the first threshold value, and when there is no target ratio greater than the first threshold value, the column type of the target resume is determined to be multi-column. In this way, the column type is determined in combination with the number of projected line segments and the first threshold value, which avoids the error of identifying the column type directly based on the number of projected line segments when there are slight intervals between the projections of the text boxes in the target resume on the X-axis but the column type is a single column, thereby improving the accuracy of determining the column type to a certain extent and further extracting the efficiency of the resume content.

[0121] Figure 7 A flowchart of a method for determining a column reading order according to an embodiment of the present application is shown, which specifically includes the following steps:

[0122] Step 702: When the column type is multiple columns, for any one of the at least one target ratio, determine the target interval distance corresponding to the target ratio.

[0123] Step 704: Determine the midline of the line segment corresponding to the target spacing distance as the column dividing reference line.

[0124] Step 706: Divide at least one text box into multiple columns according to the column division baselines, and determine the reading order between the columns.

[0125] Specifically, the target interval distance is the interval distance corresponding to the target ratio. The median line is also called the perpendicular median line, that is, the perpendicular bisector, which refers to a straight line passing through the midpoint of the line segment corresponding to the target interval distance and perpendicular to the line segment. The column reference line refers to the reference line that divides the column.

[0126] In practical applications, when the column type is multiple columns, that is, when there is a target ratio, for any target ratio, the interval distance corresponding to the target ratio is determined as the target interval distance, and then the line segment corresponding to the target interval distance is determined, and the midline of the line segment is determined as a column baseline of the target resume or each text box. One target ratio corresponds to one column baseline. After all column baselines are determined, at least one text box is divided into multiple columns according to each column baseline, that is, the number of columns of the target resume and the text boxes in each column are determined, and then the reading order between columns is determined according to the number of columns.

[0127] In addition, when determining the column dividing baseline, any vertical line on the line segment corresponding to the target spacing distance can be used as the column dividing baseline. This application only uses "determining the midline of the line segment corresponding to the target spacing distance as the column dividing baseline" as a preferred example for explanation.

[0128] After determining the target ratio, you can also determine the number of columns or columns of the target resume based on the number of target ratios, that is, the number of columns of the target resume is the sum of the number of target ratios plus one.

[0129] See also Figure 3E From the above figure, three projection line segments are obtained. The coordinates of the starting point and the end point of each projection line segment are shown in Table 2. The calculation shows that the first ratio is 0.5 / 15=1 / 30, and the second ratio is 1 / 15. Assuming that the first threshold is 1 / 20, then 1 / 15 is greater than 1 / 20, that is, 1 / 15 is the target ratio. The line segment corresponding to the target ratio 1 / 15 is 9≤x≤10 and y=0, then the midline of the line segment is x=9.5, and x=9.5 is the column reference line. Since there is only one target ratio, the target resume is a double column, and the column reference line x=9.5 divides at least one text box into two columns: text box K1, text box K2, text box K4, text box K5, text box K7, text box K8 and text box K9 belong to the first column, and text box K3, text box K6, text box K10 and text box K11 belong to the second column. Since there are two columns, the order of reading between columns is to read the text box in the first column first, and then read the text box in the second column.

[0130] It should be noted that step 702 to step 706 are a specific implementation method for determining the reading order between columns before step 208.

[0131] In the embodiment of the present application, when there are multiple projected line segments, the column division baseline is determined according to the target interval distance corresponding to each target ratio, and then based on each column division baseline, at least one text box is divided into multiple columns, and the reading order between columns is determined. In this way, the midline of the line segment corresponding to the target interval distance is determined as the column division baseline, which can quickly and effectively improve the efficiency of determining the column division baseline, and then when at least one text box is divided into multiple columns based on each column division baseline, the accuracy and speed of the division are improved, and the efficiency of the reading order between columns is improved to a certain extent, and the efficiency of extracting the content of the resume is further improved.

[0132] Figure 8 A flowchart of another method for determining a target reading order according to an embodiment of the present application is shown, which specifically includes the following steps:

[0133] Step 802: Obtain a target resume and extract at least one text box in the target resume.

[0134] Step 804: Based on at least one text box, identify the column type of the target resume.

[0135] Step 806: When the column type is a single column, the reading order within the column is determined as the target reading order of each text box.

[0136] Step 808: Based on the target reading order, read the text content from each text box in turn to obtain the resume content of the target resume.

[0137] Figure 8 Steps 802, 804, 808 and Figure 2 Steps 202, 204, and 208 are the same and will not be repeated here.

[0138] Specifically, the reading order within a column refers to the reading order of all text boxes in each column, for example, from top to bottom, from the first row of text boxes to the last row of text boxes. It should be noted that when there are multiple text boxes in a row of a column, the reading order of the text boxes in the row can be from left to right. Preferably, the reading order within a column is first from left to right and then from top to bottom. The target reading order refers to the final reading order of the text boxes determined by combining the reading order between columns and the reading order within a column.

[0139] In actual applications, when it is recognized that the column type of the target resume is a single column, the preset reading order within the column is obtained, and then the reading order within the column is determined as the target reading order of each text box.

[0140] For example, the reading order within the column is from left to right and then from top to bottom. Assuming that the target resume has three lines, the first line has two text boxes: text box c11 and text box c12, the second line has one text box text box c21, and the third line has two text boxes: text box c31 and text box c32. Then the reading order within the column is, that is, the target reading order of each text box is: text box c11, text box c12, text box c21, text box c31, text box c32.

[0141] It should be noted that the reading order within the column can be set according to demand, and this application does not limit this.

[0142] In the above implementation, at least one text box in the target resume is extracted; then based on the at least one text box, the column type of the target resume is identified; and when the column type is a single column, the reading order in the column is determined as the target reading order of each text box; then based on the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume. Using at least one text box to identify the column type of the target resume can effectively improve the recognition rate of the column type; when the column type is a single column, the target reading order of each text box is determined based on the reading order in the column, avoiding the confusion of directly extracting the resume content, which not only improves the orderliness and efficiency of extracting the resume content, but also improves the accuracy of the resume content.

[0143] Fig. 9 A flowchart of a text content reading method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0144] Step 902: Perform horizontal calibration processing on at least one text box in each column.

[0145] Step 904: According to the calibration results of each column and the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume.

[0146] Specifically, horizontal calibration refers to the process of aligning multiple text boxes into one row, where the text contents are on the same row but due to problems with text box extraction or target resume settings, multiple text boxes corresponding to the text contents are not on the same row; the calibration result refers to the result of the horizontal calibration of the text boxes.

[0147] In practical applications, the text content in the text box is extracted sequentially. Since different fields in the row are of different heights, there may be a misalignment in the horizontal direction, which is not conducive to machine reading, so horizontal calibration is required. The text boxes in any column can be processed horizontally: horizontal processing can be performed based on the coordinates of each text box; horizontal processing can be performed based on the attribute information of each text box; and horizontal processing can also be performed based on the text content of each text box, which is not limited in this application. Furthermore, according to the calibration results of each column and the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume.

[0148] It should be noted that step 902 to step 904 are a specific implementation of step 208 above.

[0149] In the embodiment of the present application, by performing horizontal calibration on at least one text box, the problem of disordered extraction caused by uneven text boxes is avoided, and then the text content is read from each text box in turn according to the calibration results of each column and the target reading order, which greatly improves the accuracy of the resume content. In addition, the horizontal calibration process is performed on the text boxes of each column, rather than on all text boxes of the target resume, which avoids the horizontal calibration process of text boxes that do not belong to the same column, reduces the processing volume of horizontal calibration of irrelevant text boxes, improves the efficiency of horizontal calibration, and greatly improves the efficiency of extracting resume content.

[0150] Fig.10 A flow chart of a lateral calibration method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0151] It is worth noting that each text box in at least one text box contains a line of text content.

[0152] Step 1002: Determine the overlapping height of the first text box and the second text box in the arrangement direction of the text lines in the target resume, wherein the first text box and the second text box are any two adjacent text boxes in the same column.

[0153] Step 1004: when the overlapping height is greater than or equal to the second threshold, determine that the first text box and the second text box are located in the same row.

[0154] Specifically, the overlapping height refers to the height of the overlapping coordinates of two adjacent text boxes on the Y axis; the second threshold is a preset value used to measure whether the overlapping height indicates that the two adjacent text boxes are in the same row, and can be adjusted according to needs.

[0155] In actual applications, for any column in each column, first determine whether any two adjacent text boxes in the column belong to the same row, that is, determine whether the first text box and the second text box are in the same row: first obtain any two adjacent text boxes in the same column and the coordinates of the two text boxes on the Y-axis, and determine the text box with the larger coordinate on the Y-axis as the first text box, that is, determine the higher text box of the two text boxes as the first text box, and determine the other text box of the two text boxes as the second text box. Then, according to the coordinates of the first text box and the second text box on the Y axis, the overlapping height of the first text box and the second text box is determined, that is, the coordinates of the upper left corner of the second text box on the Y axis are first compared with the coordinates of the lower left corner of the first text box on the Y axis. If the coordinates of the upper left corner of the second text box on the Y axis are less than or equal to the coordinates of the lower left corner of the first text box on the Y axis, the overlapping height is 0. If the coordinates of the upper left corner of the second text box on the Y axis are greater than the coordinates of the lower left corner of the first text box on the Y axis, the coordinates of the upper left corner of the second text box on the Y axis are subtracted from the coordinates of the lower left corner of the first text box on the Y axis to obtain the overlapping height. Then, the overlapping height is compared with the second threshold value. When the overlapping height is greater than or equal to the second threshold value, it is determined that the first text box and the second text box are in the same row, that is, the first text box and the second text box are listed as the same row.

[0156] Preferably, the coordinates on the Y axis of the text box are [s, e], where s represents the vertical coordinate of the lower left corner of the current text box, and e represents the vertical coordinate of the upper left corner of the current text box.

[0157] For example, the coordinate of the first text box on the Y axis is [5.2, 6.2], and the coordinate of the second text box on the Y axis is [4, 5]. Since 5.2 is greater than 5, the overlapping height of the first text box and the second text box is 0. Then the overlapping height is compared with the second threshold to determine whether the first text box and the second text box are in the same row.

[0158] For another example, the coordinate of the first text box on the Y axis is [5.2, 6.2], and the coordinate of the second text box on the Y axis is [5, 6]. Since 6 is greater than 5.2, the overlapping height of the first text box and the second text box is 6-5.2=0.8. When the second threshold is less than or equal to 0.8, the first text box and the second text box are located in the same row, that is, the first text box and the second text box are listed in the same row.

[0159] It should be noted that step 1002 to step 1004 are a specific implementation of step 902 above.

[0160] In the embodiment of the present application, by determining the overlapping height of the first text box and the second text box, when the overlapping height is greater than or equal to the second threshold, it is determined that the first text box and the second text box are located in the same row. Horizontal calibration is performed based on the overlapping height and the second threshold, which greatly improves the accuracy of horizontal calibration, thereby improving the accuracy of the extracted resume content.

[0161] Fig.11 A flow chart of another lateral calibration method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0162] Step 1102: Determine the overlapping height of the first text box and the second text box in the arrangement direction of the text lines in the target resume, wherein the first text box and the second text box are any two adjacent text boxes in the same column.

[0163] Step 1104: when the overlapping height is less than the second threshold, determine that the first text box and the second text box are located in different rows.

[0164] Specifically, the overlapping height refers to the height of the overlapping coordinates of two adjacent text boxes on the Y axis; the second threshold is a preset value used to measure whether the overlapping height indicates that the two adjacent text boxes are in the same row, and can be adjusted according to needs.

[0165] In practical applications, for any column in each column, first determine whether any two adjacent text boxes in the column belong to the same row, that is, determine whether the first text box and the second text box are located in the same row. First obtain any two adjacent text boxes in the same column and the coordinates of the two text boxes on the Y axis, and determine the text box with the larger coordinate on the Y axis as the first text box, that is, determine the higher text box of the two text boxes as the first text box, and determine the other text box of the two text boxes as the second text box. Then, according to the coordinates of the first text box and the second text box on the Y axis, determine the overlapping height of the first text box and the second text box, that is, first compare the coordinates of the upper left corner of the second text box on the Y axis with the coordinates of the lower left corner of the second text box on the Y axis. If the coordinates of the upper left corner of the second text box on the Y axis are less than or equal to the coordinates of the lower left corner of the second text box on the Y axis, the overlapping height is 0. If the coordinates of the upper left corner of the second text box on the Y axis are greater than the coordinates of the lower left corner of the second text box on the Y axis, subtract the coordinates of the upper left corner of the second text box on the Y axis from the coordinates of the lower left corner of the second text box on the Y axis to obtain the overlapping height. Then, the overlapping height is compared with the second threshold. When the overlapping height is less than the second threshold, it is determined that the first text box and the second text box are located in different rows, that is, the first text box and the second text box are listed in two rows.

[0166] Preferably, the coordinates on the Y axis of the text box are [s, e], where s represents the vertical coordinate of the lower left corner of the current text box, and e represents the vertical coordinate of the upper left corner of the current text box.

[0167] For example, the coordinate of the first text box on the Y axis is [5, 6], and the coordinate of the second text box on the Y axis is [5.8, 6.8], then the overlapping height of the first text box and the second text box is 6-5.8=0.2, and when the second threshold is greater than 0.2, the first text box and the second text box are located in different rows, that is, the first text box and the second text box are listed in two rows.

[0168] It should be noted that Fig.11 Step 1102 and Fig.10 The same as step 1002 in the above description; step 1102-step 1104 is a specific implementation of the above step 902.

[0169] In the embodiment of the present application, by determining the overlapping height of the first text box and the second text box, when the overlapping height is less than the second threshold, it is determined that the first text box and the second text box are located at different positions. The horizontal calibration is performed based on the overlapping height and the second threshold, which greatly improves the accuracy of the horizontal calibration, thereby improving the accuracy of the extracted resume content.

[0170] Fig.12 A flow chart of a method for obtaining a target resume according to an embodiment of the present application is shown, which specifically includes the following steps:

[0171] Step 1202: Obtain an initial resume and identify the text format of the initial resume.

[0172] Step 1204: When the text format of the initial resume is not a picture format, the text format of the initial resume is converted into a picture format to obtain a target resume.

[0173] Specifically, the initial resume refers to the resume obtained directly; the text format can be any specified format, such as document format (DOC, Document), text format (txt), image format, portable document format (PDF), etc.; the picture format includes image format and portable document format.

[0174] In actual applications, there are many ways to obtain the initial resume. For example, a user may send a target resume acquisition instruction to the execution subject, and accordingly, the execution subject starts to acquire the target resume after receiving the acquisition instruction; or the execution subject may automatically acquire the target resume to be extracted at a preset time interval. For example, after a preset time, a resume content extraction function may be generated. serverAutomatically obtain the target resume in the designated access area; or after a preset time, the terminal with resume content extraction function automatically obtains the target resume stored locally. This manual does not limit the method of obtaining the target resume.

[0175] Furthermore, based on the acquisition of the initial resume, the execution subject identifies the text format of the initial resume, and then compares the text format of the initial resume with the picture format (image format or portable document format). If the text format of the initial resume is any one of the image format and the portable document format, the initial resume is directly determined as the target resume; if the text format of the initial resume is neither an image format nor a portable document format, the text format of the initial resume is a non-picture format, and the text format of the initial resume needs to be converted into a picture format to obtain the target resume.

[0176] For example, when the execution subject identifies that the text format of the initial resume is DOC format, the initial resume is converted into PDF format to obtain a target resume in PDF format. For another example, when the execution subject identifies that the text format of the initial resume is image format, the initial resume is determined as the target resume.

[0177] It should be noted that step 1202 to step 1204 are a specific implementation of step 202 above.

[0178] In an embodiment of the present application, an initial resume is obtained, and then it is determined whether the text format of the initial resume is in a picture format. If so, it means that the text format of the initial resume is standardized and can be directly used as a target resume. If not, it means that the text format of the initial resume is not standardized. Therefore, it is necessary to convert the text format of the initial resume into a picture format, and determine the converted initial resume as the target resume. This can avoid the impact of the non-standard format on the extracted text box, effectively avoid the disorder problem of resume content extraction, and ensure the extraction accuracy of the text box, thereby improving the accuracy of the resume content.

[0179] The following combination Fig.13 , the resume content extraction method is further described. Among them, Fig.13 A processing flow chart of a resume content extraction method provided by an embodiment of the present application is shown, which specifically includes the following steps:

[0180] Step 1302: Obtain an initial resume and identify the text format of the initial resume.

[0181] Step 1304: When the text format of the initial resume is not a picture format, the text format of the initial resume is converted into a picture format to obtain a target resume.

[0182] Step 1306: When the text format of the initial resume is in picture format, the initial resume is determined as the target resume.

[0183] Step 1308: Extract at least one text box in the target resume.

[0184] Step 1310: Establish a plane coordinate system in the target resume, where the X-axis of the plane coordinate system is the extension direction of the text lines in the target resume, and the Y-axis is the arrangement direction of the text lines in the target resume.

[0185] Step 1312: Project at least one text box onto the X-axis to obtain at least one projection line segment of the at least one text box on the X-axis.

[0186] Step 1314: When the number of at least one projection line segment is one, determine that the column type of the target resume is single column.

[0187] Step 1316: When there are multiple projection line segments, determine the total distance between the starting point of the first projection line segment and the ending point of the last projection line segment.

[0188] Step 1318: Determine the interval distance between two adjacent projection line segments in at least one projection line segment, and respectively determine the ratio of each interval distance to the total distance.

[0189] Step 1320: When there is no target ratio among the ratios, determine that the column type of the target resume is a single column, and the target ratio is a ratio greater than the first threshold.

[0190] Step 1322: When there is a target ratio among the ratios, determine that the column type of the target resume is multi-column, and the target ratio is a ratio greater than the first threshold.

[0191] Step 1324: When the column type is a single column, the reading order within the column is determined as the target reading order of each text box.

[0192] Step 1326: When the column type is multiple columns, for any one of the at least one target ratio, determine the target interval distance corresponding to the target ratio.

[0193] Step 1328: Determine the midline of the line segment corresponding to the target spacing distance as the column dividing baseline.

[0194] Step 1330: Divide at least one text box into multiple columns according to the column division baselines, and determine the reading order between the columns.

[0195] Step 1332: Determine the target reading order of each text box according to the inter-column reading order and the preset intra-column reading order.

[0196] Step 1334: Determine the overlapping height of the first text box and the second text box on the Y axis, the first text box and the second text box being any two adjacent text boxes in the same column.

[0197] Step 1336: Perform horizontal calibration processing on at least one text box in each column.

[0198] Optionally, performing horizontal calibration processing on at least one text box in each column includes:

[0199] When the overlapping height is greater than or equal to the second threshold, determining that the first text box and the second text box are located in the same row;

[0200] When the overlapping height is less than the second threshold, it is determined that the first text box and the second text box are located in different rows.

[0201] Step 1338: According to the calibration results of each column and the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume.

[0202] The resume content extraction method provided by the present application extracts at least one text box in the target resume; then identifies the column type of the target resume based on the at least one text box; and when the column type is multiple columns, determines the target reading order of each text box according to the preset inter-column reading order and intra-column reading order; and then reads the text content from each text box in turn based on the target reading order to obtain the resume content of the target resume. Using at least one text box to identify the column type of the target resume can effectively improve the recognition rate of the column type; when the column type is multiple columns, the target reading order of each text box is determined based on the inter-column reading order and the intra-column reading order, avoiding the confusion of directly extracting the resume content, which not only improves the orderliness and efficiency of extracting the resume content, but also improves the accuracy of the resume content.

[0203] In addition, by horizontally calibrating at least one text box, the problem of disordered extraction caused by uneven text boxes is avoided, and the accuracy of the resume content is greatly improved; and by converting the initial resume in a non-image format into a target resume in an image format, it can effectively avoid the problem of disordered extraction of resume content and ensure the accuracy of text box extraction, thereby improving the accuracy of the resume content.

[0204] Corresponding to the above method embodiment, the present application also provides a resume content extraction device embodiment, Fig.14 FIG. 1 is a schematic diagram showing the structure of a resume content extraction device provided by an embodiment of the present application. Fig.14 As shown, the device comprises:

[0205] The acquisition module 1402 is configured to acquire a target resume and extract at least one text box in the target resume;

[0206] The identification module 1404 is configured to identify the column type of the target resume based on at least one text box;

[0207] The first determination module 1406 is configured to determine the target reading order of each text box according to a preset inter-column reading order and an intra-column reading order when the column type is multiple columns;

[0208] The reading module 1408 is configured to read the text content from each text box in turn based on the target reading order to obtain the resume content of the target resume.

[0209] Optionally, the identification module 1404 is further configured to:

[0210] Establish a plane coordinate system in the target resume, wherein the X-axis of the plane coordinate system is the extension direction of the text line in the target resume, and the Y-axis is the arrangement direction of the text line in the target resume;

[0211] Projecting at least one text box onto the X-axis to obtain at least one projection line segment of the at least one text box on the X-axis;

[0212] According to at least one projected line segment, the column type of the target resume is identified.

[0213] Optionally, the identification module 1404 is further configured to:

[0214] When the number of at least one projection line segment is one, determining the column type of the target resume as a single column;

[0215] When the number of at least one projection line segment is plural, it is determined that the column type of the target resume is multi-column.

[0216] Optionally, the identification module 1404 is further configured to:

[0217] In the case where the number of at least one projection line segment is plural, determining a total distance between a starting point of a first projection line segment and an ending point of a last projection line segment;

[0218] Determine the interval distance between two adjacent projection line segments in at least one projection line segment, and respectively determine the ratio of each interval distance to the total distance;

[0219] When there is a target ratio among the ratios, the column type of the target resume is determined to be multi-column, wherein the target ratio is a ratio greater than a first threshold value.

[0220] Optionally, the identification module 1404 is further configured to:

[0221] When there is no target ratio in each ratio, the column type of the target resume is determined to be single column.

[0222] Optionally, the device further includes a second determining module configured to:

[0223] In the case where the column type is multiple columns, for any one of the at least one target ratio, determining a target interval distance corresponding to the target ratio;

[0224] The midline of the line segment corresponding to the target interval distance is determined as the column dividing baseline;

[0225] Divide at least one text box into multiple columns based on the column reference lines and determine the reading order between the columns.

[0226] Optionally, the device further includes a third determining module, configured to:

[0227] When the column type is single column, the reading order within the column is determined as the target reading order of each text box.

[0228] Optionally, the reading module 1408 is configured to:

[0229] Performing horizontal calibration processing on at least one text box in each column;

[0230] According to the calibration results of each column and the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume.

[0231] Optionally, each text box in at least one text box contains a line of text content;

[0232] The reading module 1408 is configured to:

[0233] Determine an overlapping height of a first text box and a second text box in an arrangement direction of text lines in a target resume, wherein the first text box and the second text box are any two adjacent text boxes in the same column;

[0234] When the overlapping height is greater than or equal to the second threshold, it is determined that the first text box and the second text box are located in the same row.

[0235] Optionally, the reading module 1408 is configured to:

[0236] When the overlapping height is less than the second threshold, it is determined that the first text box and the second text box are located in different rows.

[0237] Optionally, the acquisition module 1402 is configured to:

[0238] Obtaining an initial resume and identifying a text format of the initial resume;

[0239] When the text format of the initial resume is not a picture format, the text format of the initial resume is converted into a picture format to obtain a target resume.

[0240] The resume content extraction device provided by the present application extracts at least one text box in the target resume; then identifies the column type of the target resume based on the at least one text box; and when the column type is multiple columns, determines the target reading order of each text box according to the preset inter-column reading order and intra-column reading order; and then reads the text content from each text box in turn based on the target reading order to obtain the resume content of the target resume. Using at least one text box to identify the column type of the target resume can effectively improve the recognition rate of the column type; when the column type is multiple columns, the target reading order of each text box is determined based on the inter-column reading order and the intra-column reading order, avoiding the confusion of directly extracting the resume content, which not only improves the orderliness and efficiency of extracting the resume content, but also improves the accuracy of the resume content.

[0241] The above is a schematic scheme of a resume content extraction device of the present embodiment. It should be noted that the technical scheme of the resume content extraction device and the technical scheme of the resume content extraction method mentioned above belong to the same concept. For details not described in detail in the technical scheme of the resume content extraction device, please refer to the description of the technical scheme of the resume content extraction method mentioned above. In addition, each component in the device embodiment should be understood as a functional module that must be established to implement each step of the program flow or each step of the method, and each functional module is not an actual functional division or separation definition. The device claim defined by such a group of functional modules should be understood as a functional module architecture that mainly implements the solution through the computer program recorded in the specification, and should not be understood as a physical device that mainly implements the solution through hardware.

[0242] Fig.15 The block diagram of a computing device 1500 provided according to an embodiment of the present application is shown. The components of the computing device 1500 include but are not limited to a memory 1510 and a processor 1520. The processor 1520 is connected to the memory 1510 via a bus 1530, and the database 1550 is used to store data.

[0243] The computing device 1500 also includes an access device 1540, which enables the computing device 1500 to communicate via one or more networks 1560. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 1540 may include one or more of any type of network interface (e.g., a network interface card (NIC)) of wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0244] In one embodiment of the present application, the above components of the computing device 1500 and Fig.15 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Fig.15 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of the present application. Those skilled in the art may add or replace other components as needed.

[0245] Computing device 1500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. Computing device 1500 may also be a mobile or stationary server.

[0246] The processor 1520 is used to execute computer executable instructions of the resume content extraction method.

[0247] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the resume content extraction method described above are of the same concept, and the details not described in detail in the technical scheme of the computing device can be found in the description of the technical scheme of the resume content extraction method described above.

[0248] An embodiment of the present application also provides a computer-readable storage medium storing computer instructions, which are used in a resume content extraction method when executed by a processor.

[0249] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the resume content extraction method described above are of the same concept, and the details not described in detail in the technical scheme of the storage medium can be found in the description of the technical scheme of the resume content extraction method described above.

[0250] An embodiment of the present application discloses a chip storing computer instructions, which, when executed by a processor, implement the steps of the resume content extraction method as described above.

[0251] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0252] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0253] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0254] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0255] The preferred embodiments of the present application disclosed above are only used to help explain the present application. The optional embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the present application. The present application selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can understand and use the present application well. The present application is only limited by the claims and their full scope and equivalents.

Claims

1. A resume content extraction method, characterized in that: include: Obtain a target resume, and extract at least one text box in the target resume; Based on the at least one text box, identifying a column type of the target resume; In the case where the column type is multiple columns, a target reading order of each text box is determined according to a preset inter-column reading order and an intra-column reading order; Based on the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume.

2. The method according to claim 1, characterized in that The step of identifying the column type of the target resume based on the at least one text box includes: Establishing a plane coordinate system in the target resume, wherein the X-axis of the plane coordinate system is the extension direction of the text lines in the target resume, and the Y-axis is the arrangement direction of the text lines in the target resume; Projecting the at least one text box onto the X-axis to obtain at least one projection line segment of the at least one text box on the X-axis; According to the at least one projection line segment, the column type of the target resume is identified.

3. The method according to claim 2, characterized in that The step of identifying the column type of the target resume according to the at least one projection line segment includes: When the number of the at least one projection line segment is one, determining that the column type of the target resume is a single column; When the number of the at least one projection line segment is multiple, it is determined that the column type of the target resume is multiple columns.

4. The method according to claim 3, characterized in that: When the number of the at least one projection line segment is plural, determining the column type of the target resume as multiple columns includes: In the case that there are multiple projection line segments, determining a total distance between a starting point of a first projection line segment and an ending point of a last projection line segment; Determine the interval distance between two adjacent projection line segments in the at least one projection line segment, and respectively determine the ratio of each interval distance to the total distance; When there is a target ratio among the ratios, the column type of the target resume is determined to be multi-column, wherein the target ratio is a ratio greater than a first threshold value.

5. The method according to claim 4, characterized in that Also includes: When there is no target ratio in each ratio, the column type of the target resume is determined to be a single column.

6. The method according to claim 4, characterized in that Before determining the target reading order of each text box according to the preset inter-column reading order and intra-column reading order, the method further includes: In the case where the column type is multiple columns, for any one of the at least one target ratio, determining a target interval distance corresponding to the target ratio; Determine the midline of the line segment corresponding to the target interval distance as the column dividing reference line; The at least one text box is divided into a plurality of columns according to the column division reference lines, and a reading order between the columns is determined.

7. The method according to claim 1, characterized in that Before the text content is read from each text box in turn based on the target reading order to obtain the resume content of the target resume, the method further includes: When the column type is a single column, the reading order within the column is determined as the target reading order of each text box.

8. The method according to any one of claims 1 to 7, characterized in that: The method of reading text content from each text box in sequence based on the target reading order to obtain the resume content of the target resume includes: Performing horizontal calibration processing on at least one text box in each column; According to the calibration results of each column and the target reading order, the text content is read from each text box in turn to obtain the resume content of the target resume.

9. The method according to claim 8, characterized in that Each text box in the at least one text box contains a line of text content; The step of performing horizontal calibration processing on at least one text box in each column comprises: Determine an overlapping height of a first text box and a second text box in an arrangement direction of text lines in the target resume, wherein the first text box and the second text box are any two adjacent text boxes in the same column; When the overlapping height is greater than or equal to a second threshold, it is determined that the first text box and the second text box are located in the same row.

10. The method according to claim 9, characterized in that Also includes: When the overlapping height is less than a second threshold, it is determined that the first text box and the second text box are located in different rows.

11. The method according to claim 1, characterized in that: The obtaining of the target resume includes: Obtaining an initial resume and identifying a text format of the initial resume; When the text format of the initial resume is not a picture format, the text format of the initial resume is converted into a picture format to obtain the target resume.

12. A resume content extraction device, characterized in that: include: An acquisition module is configured to acquire a target resume and extract at least one text box in the target resume; an identification module configured to identify the column type of the target resume based on the at least one text box; A first determination module is configured to determine a target reading order of each text box according to a preset inter-column reading order and an intra-column reading order when the column type is multiple columns; The reading module is configured to read the text content from each text box in turn based on the target reading order to obtain the resume content of the target resume.

13. A computing device, characterized in that: include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the steps of the resume content extraction method described in any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that: It stores computer instructions, which, when executed by a processor, implement the steps of the resume content extraction method described in any one of claims 1 to 11.

Images