KR102650206B1 - Server, method, and program for recommending journal submissions with high paper similarity - Google Patents

Abstract

According to an embodiment of the present invention, a server is launched to recommend journals with high paper similarity. The first degree of similarity is calculated based on the subject matter of papers covered by multiple submission journals and the content of the target paper the user wishes to submit, and the number of times papers classified by submission journals have been viewed, cited, and cited during a preset period. Based on the number of papers, the first impact factor for each submitted journal is calculated, and the success rate of submission to appropriate journals for target papers is calculated based on the paper publication rate and time required for publication by paper topic in each submitted journal. 1 Based on similarity, first impact factor, and success rate of submission to suitable journals, recommended journal information to which to submit target papers can be generated.

Description

Server, method, and program for recommending journal submissions with high paper similarity {SERVER, METHOD, AND PROGRAM FOR RECOMMENDING JOURNAL SUBMISSIONS WITH HIGH PAPER SIMILARITY}

The present invention relates to a server that provides a function to recommend submission journals with high similarity to the target paper.

Finding a suitable journal is a very important process when submitting a paper.

Despite this importance, the technology to select the optimal journal by conducting various analyzes based on objective data in the process of selecting a journal to submit a paper is not currently available to the public.

Additionally, when searching for documents such as papers, searching documents based on similarity and providing information is one of the very important functions.

Although these similarity-based search results and document recommendation functions are very important, they still only provide basic search results.

For example, it is expected that more recommended information can be provided by providing various information using similarity search results, but such technology is not currently available to the public.

Republic of Korea Patent Publication No. 10-2022-0134695, (October 5, 2022) Republic of Korea Patent Publication No. 10-2011-0122013, (November 9, 2011) Republic of Korea Patent Publication No. 10-1442719, (September 15, 2014) Republic of Korea Patent Publication No. 10-1625863, (May 25, 2016)

The purpose of the present invention is to provide a server that recommends journals with high paper similarity.

In addition, the present invention provides journal recommendation information to which the user's target paper should be submitted based on the similarity between the target paper to which the user wishes to submit and the submission journal, the impact factor for each submission journal, and the success rate of submission to suitable journals for the submission journal. It is for a different purpose.

In addition, the present invention seeks to provide a service that stores highlights displayed on an electronic document on a user terminal, annotations and memos written in the electronic document, and displays them even if the electronic document is viewed by another user.

One aspect of a server that recommends submission journals with high paper similarity according to an embodiment of the present invention for achieving the above object includes: a storage unit storing information for each submission journal and at least one instruction; and a processor executing the at least one instruction, wherein the processor calculates a first degree of similarity based on the subject matter of papers covered by a plurality of submission journals and the content of the target paper to which the user wishes to submit, and The first impact factor for each submitted journal is calculated based on the number of times the classified papers are viewed, cited, and cited during a preset period, and the time required to publish the paper by topic for each submitted journal and the paper Based on the publication rate, the success rate of submission to a suitable journal for the target paper can be calculated, and journal recommendation information to which the target paper will be submitted can be generated based on the calculated first similarity, first influence factor, and submission success rate.

In addition, the processor classifies papers classified by submission journal into a plurality of preset paper topics, and uses an artificial intelligence model to determine the number of searches for each period classified by preset period for papers classified by paper subject, Based on the number of views, number of citations, and number of cited papers, the second impact factor is calculated for each period of time for each paper subject of each submitted journal, and based on the calculated second impact factor, each paper subject of each submitted journal is calculated. An influence index trend can be generated, and based on the calculated influence index trend using the artificial intelligence model, the expected influence score of the target paper can be calculated when the target paper is published in each submitted journal.

In addition, the processor calculates a second similarity between the paper published in each submission journal and the target paper, the first similarity, the first impact index, the success rate of submission to a suitable journal, the expected impact score of the target paper, and the Based on the second similarity, journal recommendation information for submitting the target paper is generated, and if the second similarity falls within a preset range, the first content in the similar paper that is judged to be problematic for publication of the target paper Extract the second content in the target paper corresponding to the first content, and generate a prompt for input into the generative model based on the subject of the target paper, the first content, and the second content. and inputting the generated prompt into the generative model to determine whether the similarity problem with the first content can be solved without changing the meaning of the second content, and if it is determined that the problem is solvable, about the second content. You may request to print at least one example change.

In addition, it further includes a communication unit for communicating with a user terminal connected to the server, and at least one second paper for providing recommended information to the user based on similarity to the contents of at least one first paper viewed by the user. Select a paper, and among the contents confirmed by browsing the first paper using an artificial intelligence model, at least one search condition based on the user's search for the first paper and content related to the topic of the paper the user is searching for Generate a first keyword, calculate a recommendation score for each of the second papers based on the generated at least one first keyword, and recommend the at least one second paper based on the calculated recommendation score. You can.

In addition, using an artificial intelligence model, a plurality of second keywords for the user are selected based on the user's paper search conditions and contents corresponding to the positions of highlights and notes displayed in papers viewed by the user, An interest level value is calculated and set for each of the plurality of selected second keywords, and when a new paper is viewed by the user, the content related to the second keyword among the content included in the new paper is calculated based on the set interest level value. Calculate the matching degree, output a visual effect to inform the user at the location corresponding to the content that is higher than the preset matching degree, and recall whether the user highlighted or wrote a memo on the content that is higher than the preset matching degree. It can be learned by inputting it into an artificial intelligence model.

In addition, one aspect of the artificial intelligence-based submission journal recommendation method according to an embodiment of the present invention to achieve the above object is that the processor calculates the first similarity based on the content of the target paper, and the processor calculates the first similarity for each submission journal. 1 Calculate the impact factor, the processor calculates the success rate of submission to appropriate journals for each submission journal, the processor generates journal recommendation information to submit the target paper, and provides the journal recommendation information generated by the processor to the user terminal. You can.

According to an embodiment of the present invention, a server that recommends submission journals with high paper similarity can be provided.

In addition, according to another embodiment of the present invention, recommended journal information to which the user will submit the target paper is provided based on the similarity between the target paper to which the user wishes to submit and the submission journal, the impact factor for each submission journal, and the success rate of submission to suitable journals. can do.

In addition, according to another embodiment of the present invention, it is possible to provide a service that displays highlights displayed in electronic documents on a user terminal, annotations and memos written on a server rather than the electronic document, even if the electronic document is viewed by another user. You can.

Figure 1 is a schematic diagram of a system for recommending journals with high paper similarity according to an embodiment of the present invention.
Figure 2 is a block diagram of an artificial intelligence-based submission journal recommendation device according to an embodiment of the present invention.
Figure 3 is a flowchart of an artificial intelligence-based journal recommendation method according to an embodiment of the present invention.
Figure 4 is a diagram illustrating generating journal recommendation information using an artificial intelligence model and providing it to the user.
Figure 5 is a diagram illustrating the use of an artificial intelligence model to calculate the expected influence score after a target paper is published in a journal and provide it to the user.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention. While describing each drawing, similar reference numerals are used for similar components.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, a first component may be named a second component, and similarly, the second component may also be named a first component without departing from the scope of the present invention. The term and/or includes any of a plurality of related stated items or a combination of a plurality of related stated items.

When a component is said to be "connected" or "connected" to another component, it is understood that it may be directly connected to or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

The terms used in this application are only used to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless explicitly defined in the present application, should not be interpreted in an ideal or excessively formal sense. No.

Functions related to artificial intelligence according to the present disclosure are operated through a processor and memory. The processor may consist of one or multiple processors. At this time, one or more processors may be a general-purpose processor such as a CPU, AP, or DSP (Digital Signal Processor), a graphics-specific processor such as a GPU or VPU (Vision Processing Unit), or an artificial intelligence-specific processor such as an NPU. One or more processors control input data to be processed according to predefined operation rules or artificial intelligence models stored in memory. Alternatively, when one or more processors are dedicated artificial intelligence processors, the artificial intelligence dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

Predefined operation rules or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that the basic artificial intelligence model is learned using a large number of learning data by a learning algorithm, thereby creating a predefined operation rule or artificial intelligence model set to perform the desired characteristics (or purpose). It means burden. This learning may be accomplished in the device itself that performs the artificial intelligence according to the present disclosure, or may be accomplished through a separate server and/or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples described above.

An artificial intelligence model may be composed of multiple neural network layers. Each of the plurality of neural network layers has a plurality of weight values, and neural network calculation is performed through calculation between the calculation result of the previous layer and the plurality of weights. Multiple weights of multiple neural network layers can be optimized by the learning results of the artificial intelligence model. For example, during the learning process, a plurality of weights may be updated so that loss or cost values obtained from the artificial intelligence model are reduced or minimized. Artificial neural networks may include deep neural networks (DNN), such as Convolutional Neural Network (CNN), Deep Neural Network (DNN), Recurrent Neural Network (RNN), Restricted Boltzmann Machine (RBM), Deep Belief Network (DBN), Bidirectional Recurrent Deep Neural Network (BRDNN), or Deep Q-Networks, etc., but are not limited to the above examples.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings.

Figure 1 is a schematic diagram of a system 10 for recommending journals with high paper similarity according to an embodiment of the present invention.

Referring to FIG. 1, the artificial intelligence-based submission journal recommendation system 10 according to an embodiment of the present disclosure includes a submission journal recommendation server 100 and a user terminal 200.

At this time, the external server 300 can be an external paper search site or a server of an external paper submission journal.

The submission journal recommendation server 100 can provide a submission journal recommendation service through the web or an application, and can provide the submission journal recommendation service to the terminal 200 connected to the server through the communication unit 120. In addition, the submission journal recommendation server 100 may process the information stored in the storage unit 130 according to a request from the terminal 200 and provide it to the terminal 200, and various types of information may be processed in the corresponding external server 300. After searching the data, the information may be processed and provided to the terminal 200.

The terminal 200 is a communication capable desktop computer, laptop computer, laptop, smart phone, tablet PC, mobile phone, smart watch ( smart watch, smart glass, e-book reader, PMP (portable multimedia player), portable game console, navigation device, digital camera, DMB (digital multimedia broadcasting) player, digital voice recorder It may be a digital audio recorder, digital audio player, digital video recorder, digital video player, PDA (Personal Digital Assistant), etc.

The server may communicate with the terminal 200 or the external server 300 through one or more wired or wireless network connections.

Below, the artificial intelligence-based submission journal recommendation server 100, method, and program according to an embodiment of the present invention will be described in more detail with reference to other drawings.

Figure 2 is a block diagram of an artificial intelligence-based submission journal recommendation device according to an embodiment of the present invention.

Referring to FIG. 2, the artificial intelligence-based submission journal recommendation device according to an embodiment of the present invention includes a processor 110, a communication unit 120, and a storage unit 130.

Here, the at least one processor 110 is a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor 110 on which methods according to embodiments of the present invention are performed. It can mean. Each memory and storage device may be comprised of at least one of a volatile storage medium and a non-volatile storage medium.

The processor 110 typically controls the overall operation of the submission journal recommendation server 100. For example, the processor 110 may generally control the submission journal recommendation server 100 by executing programs stored in the memory. The submission journal recommendation server 100 may include at least one processor 110.

The processor 110 may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input/output operations. Commands may be provided to the processor 110 from memory, or may be received through the communication unit 120 and provided to the processor 110. For example, the processor 110 may be configured to execute instructions according to program codes stored in a recording device such as a memory.

The storage unit 130 according to an embodiment may store one or more instructions, and at least one processor 110 of the above-described server performs the operation according to an embodiment by executing the one or more instructions stored in the memory. It can be done.

In an embodiment of the present invention, the storage means may be referred to as a storage unit 130, a memory, and a database.

The storage unit 130 may be one of read only memory (ROM) and random access memory (RAM), and the storage device may be flash memory or hard disk drive (HDD). ), a solid state drive (SSD), or various memory cards (eg, micro SD card).

Additionally, the server may include a transceiver that performs communication through a wireless network. Additionally, the server may further include an input interface device, an output interface device, a storage device, etc. Each component included in the server is connected by a bus and can communicate with each other.

Examples of servers that can communicate include desktop computers, laptop computers, laptops, smart phones, tablet PCs, mobile phones, and smart watches. (smart watch), smart glass, e-book reader, PMP (portable multimedia player), portable game console, navigation device, digital camera, DMB (digital multimedia broadcasting) player, digital voice It may be a digital audio recorder, digital audio player, digital video recorder, digital video player, PDA (Personal Digital Assistant), etc.

The communication unit 120 may include one or more components that allow the electronic device to communicate with a server or external device. For example, the communication unit 120 may include a short-range communication unit 120, a mobile communication unit 120, and a broadcast reception unit.

Networks include Local Area Network (LAN), Wide Area Network (WAN), Value Added Network (VAN), Personal Area Network (PAN), and mobile radio communication. It can be implemented with various types of wired and wireless networks, such as network or satellite communication network.

A network (not shown) that communicates and connects the submission journal recommendation server 100, the terminal connected to the server 200, and the external server 300 is a Local Area Network (LAN) or a Wide Area Network (WAN). ) or a Value Added Network (VAN), etc., or any type of wireless network such as a mobile radio communication network or a satellite communication network. In addition, the network is at least two of a local area network (LAN), a wide area network (WAN), a value added network (VAN), a mobile radio communication network, or a satellite communication network. It may include a combination, and is a comprehensive data communication network that allows each network constituent shown in Figure 1 to communicate smoothly with each other, and includes wired Internet, wireless Internet, and mobile wireless communication networks. Wireless communications include, for example, wireless LAN (Wi-Fi), Bluetooth, Bluetooth low energy, ZigBee, WFD (Wi-Fi Direct), UWB (ultra-wideband), and infrared communication (IrDA, infrared data). Association), NFC (Near Field Communication), etc., but are not limited to these.

The storage unit 130 includes a flash memory type, hard disk type, multimedia card micro type, card type memory (for example, SD or XD memory, etc.), RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic It may include at least one type of storage medium among disks and optical disks. Additionally, the submission journal recommendation server 100 may operate a web storage or cloud server that performs a memory storage function on the Internet.

In one embodiment, the storage unit 130 stores information for each submission journal and at least one instruction.

At this time, information for multiple submission journals includes the number of papers published in the submission journals, type, topic, review procedure, paper publication rate, time required to publish papers, and information on users for each submission journal (e.g., age, education level, Field of employment, etc.), and any information related to the journal submitted can be applied.

Additionally, the storage unit 130 may store a pre-trained artificial intelligence model.

Additionally, the storage unit 130 may store user information and may include at least one of the user's age, school, major, degree, and workplace information.

Figure 3 is a flowchart of an artificial intelligence-based journal recommendation method according to an embodiment of the present invention.

Figure 4 is a diagram illustrating generating journal recommendation information using an artificial intelligence model and providing it to the user.

Figure 5 is a diagram illustrating the use of an artificial intelligence model to calculate the expected influence score after a target paper is published in a journal and provide it to the user.

The operation process of the artificial intelligence-based submission journal recommendation server 100 according to an embodiment of the present invention will be described with reference to FIGS. 3 to 5.

The processor 110 calculates the first similarity level based on the contents of the target paper (S100).

In one embodiment, the processor 110 may calculate the first similarity level based on the subject of papers covered by a plurality of submission journals and the contents of the target paper to which the user wishes to submit.

This is a configuration to find the submission journal that is most suitable for the topic and contents of the paper the user wishes to submit. For example, if the processor 110 determines that the submission journal perfectly matches the topic and content of the target paper, the processor 110 may calculate the first similarity as 100. Conversely, the processor 110 may calculate a first similarity close to 0 when it is determined that the submission journal has little relationship with the topic and content of the target paper.

In one embodiment, the processor 110 may receive the topic of the target paper from the user terminal 200 and receive at least one representative keyword that can represent the content of the target paper.

Additionally, the processor 110 may calculate a first degree of similarity with each submission journal based on the paper topic and representative keywords received from the terminal 200.

The processor 110 calculates the first impact index for each submitted journal (S200).

When submitting a paper, it is important to calculate the similarity between the target paper and the submitting journal as shown above, but submitting and publishing the paper considering the influence of the submitting journal can also be of great help to the user's paper and career.

Therefore, the submission journal recommendation server 100 according to an embodiment of the present invention calculates the influence factor for each submission journal and considers this along with other indicators to recommend a journal to the user to submit the paper to.

The processor 110 may calculate the first impact index for each submission journal based on the number of times the papers classified by submission journal were viewed, the number of citations, and the number of cited papers during a preset period. In addition, the processor 110 can apply any data that is helpful in calculating the impact factor of the submitting journal.

Specifically, the first influence index can be determined by Equation 1 below.

In Equation 1 above, A1 means the first influence index, L means the number of views of the link where the paper was published, V means the number of times the classified paper was actually viewed during a preset period, and C means the number of times the classified paper was actually viewed during the preset period. It refers to the number of citations of classified papers during a set period, and N refers to the number of papers published in the submission journal during a preset period.

A relatively high first impact factor means that the paper's influence within the submitting journal is relatively high, and a relatively low first impact factor means that the paper's influence within the submitting journal is relatively low.

Through this, users can predict how much influence the papers covered in the submission journal will have on the users who use the submission journal.

Additionally, the processor 110 calculates the success rate of submission to an appropriate journal for each submission journal (S300).

The processor 110 may calculate the success rate of submission to a suitable journal for the target paper based on at least one of the average time required to publish a paper by paper topic and the paper publication rate by paper topic in each submission journal.

In one embodiment, the processor 110 calculates the average time required to publish a paper by paper topic for each submission journal and the paper publication rate by paper topic, and applies a value that matches the paper topic of the target paper to a suitable journal for the target paper. The submission success rate can be calculated.

At this time, the processor 110 applies the paper publication rate of the matching paper topic, and the greater the number of papers exceeding the preset paper publication time for the paper review related to the paper topic of the corresponding submission journal, the higher the success rate of submission of the target paper to a suitable journal. It can be calculated low.

The processor 110 generates journal recommendation information to submit the target paper (S400).

Journal recommendation information generated by the processor 110 is provided to the user terminal 200 (S500).

The processor 110 generates journal recommendation information to which the target paper will be submitted based on the first similarity, first impact factor, and suitable journal submission success rate calculated through the above process.

Additionally, the processor 110 may provide the generated journal recommendation information to the user terminal 200.

At this time, the processor 110 may provide information by sorting the order of submission journals based on one of the first similarity, first influence index, or suitable journal submission success rate according to the request signal received from the user terminal 200. .

In one embodiment, the processor 110 may receive input from the user terminal 200 as a priority or weight for each of the first similarity, the first influence factor, and the success rate of submission to an appropriate journal.

For example, when a user selects a journal to publish a paper, if the influence of the submitting journal is given top priority and the first similarity factor is given top priority, the processor 110 determines the first influence factor with the highest priority. Weights are assigned, and the next highest weight is assigned to the first similarity to calculate a recommendation score for each submission journal, and based on the calculated recommendation score, submission journal recommendation information can be provided to the terminal 200.

As another example, the processor 110 may directly input and receive weight values to be assigned to each of the first similarity, the first influence factor, and the success rate of submission to an appropriate journal from the terminal 200. In this case, the processor 110 assigns the weight values received from the terminal 200 to the first similarity, the first influence index, and the success rate of submission to suitable journals, then calculates a recommendation score for each submission journal, and calculates the calculated recommendation score. Based on this, submission journal recommendation information can be provided to the terminal 200.

The processor 110 calculated the first influence index for each submission journal in S200, and in the example below, a more detailed service can be provided by calculating the second influence index for each paper topic of each submission journal.

In one embodiment, the processor 110 may classify papers classified by submission journals by a plurality of preset paper topics.

The processor 110 uses a pre-trained artificial intelligence model to determine the number of times a paper is searched, viewed, cited, and cited during each period classified by preset period for papers classified by paper topic. Based on this, the second impact index for each period for each paper topic of each submitted journal can be calculated.

Additionally, the processor 110 may generate an influence index trend for each paper topic of each submission journal based on the calculated second influence index.

For example, the processor 110 may calculate the second impact index on a yearly basis for each paper topic of each submitted journal. In addition, the processor 110 can determine the trend of increase or decrease in the second influence factor for each paper topic of each submitting journal based on the second influence factor calculated in this way.

Even if the first impact factor of the first submitting journal is calculated to be higher than that of the second submitting journal, the second submitting journal may be higher with respect to the first paper topic.

In addition, even if the second impact factors of the second and third submission journals are similar with respect to the first paper topic, the second impact factor of the second submission journal with respect to the first paper topic tends to decrease every year, and the third submission journal's second impact factor with respect to the first paper topic is similar. If the second influence factor for the subject of the first paper of the submitting journal tends to increase every year, and all other conditions are the same, the processor 110 determines whether the user's paper and the user's paper will be submitted to the third submission journal. It can be judged to be more helpful for your career.

The processor 110 provides a first message that includes at least one of the number of searches, number of views, number of citations, number of cited papers, and grade calculated for the degree of knowledge creation of the paper after the paper was published in each submitted journal. Based on the information, you can receive/obtain an influence score/grade that calculates the influence of the paper published in each submitted journal.

In addition, the artificial intelligence model can learn how to calculate the expected influence score of the paper by performing learning based on the first information and the calculated influence score/grade.

The artificial intelligence-based submission journal recommendation server 100 according to an embodiment of the present invention can provide a similarity calculation service for determining the similarity of target papers.

In one embodiment, the processor 110 may calculate a second similarity between the paper published in each submitted journal and the target paper.

The second similarity may include a plurality of previously classified ranges. For example, 0 to 10% can be classified as good, 10% to 15% as caution, 15% to 20% as very caution, and over 20% as dangerous.

And, if the calculated second similarity is within a preset range (e.g., caution, extreme caution), the processor 110 determines that it is a problem within a similar paper that is judged to be a problem in the publication of the target paper. The first content is extracted, and the second content corresponding to the first content is extracted from the target paper.

The processor 110 generates a prompt for input into the generative model based on the topic, first content, and second content of the target paper.

Then, the processor 110 inputs the generated prompt into the generative model to determine whether the similarity problem with the first content can be solved without changing the meaning of the second content, and if it is determined that the problem is solvable, the second content Ask to print at least one example change.

Through this, the processor 110 can extract a method for lowering the second similarity for the target paper using a generative model and provide this to the user to edit the target paper and then submit it.

The submission journal recommendation server 100 according to an embodiment of the present invention can identify the needs of users who search and view papers and provide paper recommendation information based on this.

In one embodiment, the processor 110 selects at least one second paper to provide recommended information to the user based on similarity to at least one first paper viewed by the user.

The processor 110 searches the first paper using an artificial intelligence model and selects at least one first keyword based on the search conditions under which the user searched the first paper and content related to the topic of the paper the user is searching. creates .

Additionally, the processor 110 may calculate a recommendation score for each second paper based on at least one first keyword and recommend at least one second paper based on the calculated recommendation score.

For example, by calculating a recommendation score for each of the plurality of searched second papers and sorting the second papers based on the order of the recommendation scores, paper recommendation information can be provided to the user. Through this, the paper the user is looking for can be prioritized. It can be provided in order of ranking.

In one embodiment, the server may use an explainable artificial intelligence model.

Specifically, the processor 110 requests the user to explain the purpose of searching the paper by inputting the search conditions under which the user searched the first paper and the generated at least one first keyword into an explainable model, and provides an explainable model. The paper search purpose can be obtained from the model.

At this time, the purpose of the paper search obtained by requesting an explainable model may be an explanation of the paper the user wishes to search.

Then, the processor 110 generates a command prompt for input into the generative model based on the paper search purpose, enters the generated command prompt to select at least one third paper that meets the user's paper search purpose, and , requesting a summary explanation of the selected third paper.

Next, the processor 110 may provide at least one third paper obtained from the generative model to the user terminal 200.

The submission journal recommendation server 100 according to an embodiment of the present invention searches and provides papers that the user wishes to search using a descriptive model and a generative model through the above process, thereby providing papers that were limited to previous search terms. By using an artificial intelligence model for search, it has the effect of being able to more accurately identify user needs and provide recommended papers.

In the past, electronic documents such as papers were called and downloaded from the main system 10 through a relay server, and then the data was provided to the user terminal 200.

However, the server according to an embodiment of the present invention can provide a service that allows electronic documents to be stream downloaded and viewed to the user terminal 200 without going through a relay server.

Additionally, in the past, there was a problem that other users could not check the electronic document by highlighting it, annotating it, or saving the written memo within the electronic document.

However, the server stores the highlights, annotations, and written notes displayed in the electronic document in the storage unit 130 of the server and provides them to the terminal 200 of other users viewing the electronic document, thereby providing the highlights, annotations, and notes. This has the effect of being able to share at least one of them with other users.

In one embodiment, the processor 110 uses an artificial intelligence model to select a plurality of second keywords for the user based on the user's paper search conditions and the contents corresponding to the positions of highlights and notes displayed in papers viewed by the user. You can select. Additionally, the processor 110 may calculate and set an interest level value for each of the plurality of selected second keywords.

For example, when the same content is repeatedly highlighted or a memo is written, the processor 110 may calculate a higher interest level value for the corresponding second keyword.

For example, when a highlight is displayed or a memo is written on the topic of a paper searched by the user or content related to the paper search conditions, the processor 110 may calculate a higher interest level value for the corresponding second keyword. there is.

Additionally, when a new paper is viewed by the user, the processor 110 calculates a degree of matching for content related to the second keyword among the content included in the new paper based on the set interest level value.

The processor 110 outputs a visual effect to inform the user at a location corresponding to content that is higher than a preset matching degree.

Additionally, the processor 110 can input and learn whether the user highlighted or wrote a memo on content that is higher than the preset matching degree by inputting it to the artificial intelligence model.

For example, if the user does not highlight or write a memo on the contents of the visual effect output, the artificial intelligence model can learn these results to reduce the interest level of the corresponding second keyword. .

In this way, the server trains the artificial intelligence model on the signals input during the user's paper search and viewing process, gradually improving the accuracy of the artificial intelligence model and viewing the content that the user is expected to highlight or write notes on. It becomes possible to predict accurately.

Methods according to the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. Computer-readable media may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on a computer-readable medium may be those specifically designed and configured for the present invention, or may be known and usable by those skilled in the art of computer software.

Examples of computer-readable media may include hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions may include machine language code such as that created by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The above-described hardware device may be configured to operate with at least one software module to perform the operations of the present invention, and vice versa.

Additionally, the above-described method or device may be implemented by combining all or part of its components or functions, or may be implemented separately.

Although the present invention has been described above with reference to preferred embodiments, those skilled in the art may make various modifications and changes to the present invention without departing from the spirit and scope of the present invention as set forth in the claims below. You will understand that you can do it.

10: Submission journal recommendation system
100: Submission journal recommendation server
110: processor
120: Department of Communications
130: storage unit
200: terminal
300: external server

Claims (5)

A communication unit that communicates with a user terminal;
a storage unit storing information for each submission journal and at least one instruction; and
Includes a processor executing the at least one instruction,
The processor,
The first degree of similarity is calculated based on the topic of papers covered by multiple submission journals and the content of the target paper the user wishes to submit,
Calculate the first impact index for each submission journal based on the number of times the papers classified by submission journal were viewed, cited, and cited during a preset period,
Calculate the success rate of submission to appropriate journals for the above-mentioned target papers based on the time taken to publish each paper by subject and the paper publication rate of each submission journal.
Generating journal recommendation information to which the target paper will be submitted based on the calculated first similarity, first influence index, and suitable journal submission success rate,
Calculate the second similarity between the paper published in each submitted journal and the target paper,
Generating journal recommendation information to which the target paper will be submitted based on the first similarity, the first impact index, the success rate of submission to a suitable journal, the expected influence score of the target paper, and the second similarity,
If the second similarity falls within a preset range,
Extract the first content from similar papers that are judged to be problematic in publishing the target paper,
Extract the second content from the target paper that corresponds to the first content,
Generating a prompt for input into a generative model based on the subject of the target paper, the first content, and the second content,
Whether the similarity problem with the first content can be solved by inputting the generated prompt into the generative model without changing the meaning of the second content, and if the problem is determined to be solvable, at least one question about the second content Characterized by requesting to output a change example of,
A server that recommends journals for submissions with high paper similarity. According to paragraph 1,
The first influence index is determined by the following equation,

In the above equation, A1 means the first influence factor, L means the number of views of the link where the paper was posted, V means the number of times the classified paper was actually viewed during a preset period, and C means the number of times the classified paper was actually viewed during the preset period. This refers to the number of citations of papers classified during the period,
A server that recommends journals for submissions with high paper similarity. According to paragraph 1,
The processor,
Papers classified by submission journal are classified by multiple pre-set paper topics,
For papers classified by paper topic using an artificial intelligence model, each submission journal's paper topic is based on the number of searches, number of views, number of citations, and number of cited papers during each period classified into preset period units. Calculate a second influence index for the period,
Based on the second influence index calculated above, the influence index trend for each paper topic of each submitted journal is generated,
Characterized in calculating the expected influence score of the target paper when the target paper is published in each submitted journal, based on the calculated influence index trend using the artificial intelligence model,
A server that recommends journals for submissions with high paper similarity. According to paragraph 3,
Selecting at least one second paper to provide recommended information to the user based on similarity to the content of at least one first paper viewed by the user,
Generating at least one first keyword based on the search conditions under which the user searched the first paper and content related to the topic of the paper the user is searching among the contents confirmed by viewing the first paper using an artificial intelligence model. do,
Calculating a recommendation score for each of the second papers based on the generated at least one first keyword,
Recommend the at least one second paper based on the calculated recommendation score,
A server that recommends journals for submissions with high paper similarity. According to clause 4,
Using an artificial intelligence model, a plurality of second keywords are selected for the user based on the user's paper search conditions and contents corresponding to the positions of highlights and notes displayed in papers viewed by the user, and the selected plurality of keywords are selected using an artificial intelligence model. Calculate and set interest levels for each of the second keywords,
When a new paper is viewed by the user, a matching degree is calculated based on the set interest value for content related to the second keyword among the content included in the new paper,
Output a visual effect to inform the user at a location corresponding to content that is higher than the preset matching degree,
Characterized in learning whether the user highlighted or wrote a memo on content that is higher than the preset matching degree by inputting it to the artificial intelligence model,
A server that recommends journals for submissions with high paper similarity.