Method, device, equipment and storage medium for automatically generating news channel information

Abstract

The present invention discloses a method, device, equipment and storage medium for automatically generating news channel information. The method comprises the following steps: obtaining source news titles of target news topics, crawling corresponding contents in time sequence of each source news title, taking source news titles corresponding to network addresses not existing in a database as target source news titles, determining basic news information from crawled contents for each target source news title, and selecting several sentences to generate news summaries of the target news topics; converting text contents of the news summaries into voice information; crawling news event pictures related to the target news topics through the network addresses of each target source news title; synthesizing target news topic videos according to text contents, voice information and news event pictures of the news summaries; the method can automatically integrate hot news currently being discussed, and can automatically process them into audio and video, thus significantly reducing the manual labor of the news industry.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a method, device, equipment, and storage medium for automatically generating news channel information. Background Art

[0002] With the development of technology, the channels and methods for people to obtain news information are gradually changing, from traditional newspapers, books, radio and television, to portal websites during the rise of the Internet and news broadcasts on social platforms. Each change is an improvement to the existing technology.

[0003] At present, the news on these channels is more about reporting the progress and status of current events, and rarely traces the causes of events to give listeners an overall understanding of the events. Although news broadcasts on social platforms have improved and relevant events can be explored, they usually rely on manual search for information, extraction of key information, and compilation into new texts, which requires a lot of manpower and is inefficient. The characteristics of high investment and low returns have led to no one doing these jobs specifically. If readers only want to simply understand the information related to the news of the day, they usually have to search for relevant content themselves. Since current search engines do not necessarily push content based on precise matches, readers still need to filter out which ones are relevant and which ones are irrelevant, wasting energy and time. Summary of the invention

[0004] The main purpose of the present invention is to provide a method, device, equipment, and storage medium for automatically generating news channel information to solve the above technical problems.

[0005] In a first aspect, the present invention provides a method for automatically generating news channel information, the method comprising the following steps:

[0006] Step S1: determine the target news topic; obtain different source news titles corresponding to the target news topic respectively, crawl the corresponding content according to the time sequence of each source news title; determine whether the network address of each source news title already exists in the database;

[0007] Step S2: taking the source news title corresponding to the network address not existing in the database as the target source news title, determining the basic news information from the crawled content for each target source news title, selecting several sentences from the crawled content to generate a news summary of the target news topic, and storing the basic news information and the news summary in the database;

[0008] Step S3: converting the text content of the news summary into voice information, and saving the language information;

[0009] Step S4: crawling news event pictures related to the target news topic through the network address of each target source news title, and storing the crawled news event pictures in the database;

[0010] Step S5: synthesizing a target news theme video according to the text content of the news summary, the voice information, and the news event picture;

[0011] Step S6: inserting the generated target news theme video into the current news, marking the title of the target news theme and the corresponding time, and generating exclusive channel information related to the target news theme.

[0012] Preferably, the method for automatically generating news channel information according to claim 1 is characterized in that in step S1, crawling the news content of the source news title in an incremental crawler manner specifically includes:

[0013] A preset web crawler tool is used, which establishes a network address set from startup to shutdown, and the network addresses crawled during the operation of the preset web crawler tool are placed in the network address set. Each time a request is initiated to a network address to be processed, the network address to be processed is compared with the existing network addresses in the network address set, and duplicate network addresses that have been crawled are filtered out.

[0014] Preferably, the step of generating a news summary of the target news topic in step S2 is:

[0015] The summary uses an extractive algorithm to calculate the weight of each sentence, and selects several sentences according to the weight of each sentence to form a paragraph as the summary of this news. The weight of each sentence is calculated by the following formula:

[0016]

[0017] Where d is the damping coefficient, S0 is the number of words repeated in the source news title and the sentence, S i is the number of words in the ith sentence, In(V i ) and Out(V j ) are pointing to the node V i The node set and slave node V i The set of nodes pointed to by the outgoing edge, WS(V i ) represents node V i rank value, WS(V j ) represents node V j rank value, w ji Indicates V j To V i The weight is the similarity between sentences, Wjk Indicates V j To V k The weight of S0 and S i The ratio of news headlines to sentence weights is used to realize the influence of news headlines on sentence weights.

[0018] Preferably, the step S4 further comprises:

[0019] Each time a current news event picture related to the target news topic is crawled, the preset field in the corresponding network address is removed and cached to obtain the network address to be compared;

[0020] When crawling the next news event image related to the target news topic, compare the difference between the two network addresses to be compared:

[0021] If the comparison result is no difference, skip the current step and continue to crawl the next news event picture; if

[0022] If the comparison result shows that there is a difference, the step of saving the crawled news event pictures in the database is executed.

[0023] Preferably, in step S5, the text content of the news summary is used as the subtitle of the target news topic, the language information is used as the audio of the target news topic, and the news event picture is used as the background of the target news topic video.

[0024] Optionally, before step S5, the method further includes:

[0025] In response to a query request input by a user, searching for a corresponding news theme video from the database, and displaying the found synthesized target news theme video;

[0026] After receiving the user's selection operation on the video, the step S6 is executed.

[0027] In a second aspect, the present invention further provides a device for automatically generating news channel information, the device comprising:

[0028] Crawler module: used to determine the target news topic; obtain different source news titles corresponding to the target news topic respectively, crawl the corresponding content according to the time sequence of each source news title; determine whether the network address of each source news title already exists in the database;

[0029] Automatic summary module: if it is determined that the target source news title corresponding to the network address does not exist in the database, then for each target source news title, basic news information is determined from the crawled content, and several sentences are selected from the crawled content to generate a news summary of the target news topic, and the basic news information and the news summary are stored in the database;

[0030] Synthetic video module: converting the text content of the news summary into voice information and saving the language information; crawling news event pictures related to the target news topic through the network address of each target source news title, and saving the crawled news event pictures in the database; synthesizing the target news topic video according to the text content of the news summary, the voice information, and the news event pictures;

[0031] The video publishing module is used to respond to the query request input by the user, search for the corresponding news theme video from the database, and display the synthesized target news theme video found; after receiving the user's selection operation on the video, the generated target news theme video is inserted into the current news, the title of the target news theme and the corresponding time are marked, and the exclusive channel information related to the target news theme is generated.

[0032] In a third aspect, the present invention further proposes a device for automatically generating news channel information, characterized in that the device comprises: a memory, a processor, and a program for automatically generating news channel information stored in the memory and executable on the processor, wherein the program is configured to implement the steps of the method for automatically generating news channel information as described in any one of the above items.

[0033] In a third aspect, the present invention further proposes a storage medium storing a program for automatically generating news channel information, wherein the program, when executed by a processor, implements the steps of the method for automatically generating news channel information as described above.

[0034] The present invention provides a method that can automatically integrate past news content summaries and can automatically broadcast them in audio and video. It can integrate the hot news currently being discussed, find relevant past information, summarize it into brief text, and process it into audio and video, which is more in line with the current fast-food way of obtaining information. Find past news based on current hot news, convert them into news summaries, and produce new audio and video. By implementing this method, the efficiency of obtaining information is improved, and the manual labor of the news industry is significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention;

[0036] Figure 2 A schematic diagram of a flow chart of a first embodiment of a method for automatically generating news channel information according to the present invention;

[0037] Figure 3 An entity relationship diagram of a database in an embodiment of the present invention;

[0038] Figure 4 It is another flow chart of an embodiment of the method for automatically generating news channel information of the present invention;

[0039] Figure 5 This is a functional module diagram of the first embodiment of the device for automatically generating news channel information according to the present invention.

[0040] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0041] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0042] Reference Figure 1 , Figure 1 The figure is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0043] like Figure 1 As shown, the automatic news channel information generation device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005; wherein the communication bus 1002 is used to realize the connection and communication between these components; the user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 1003 may optionally include a standard wired interface and a wireless interface; the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a Wi-Fi interface); the memory 1005 may be a high-speed RAM memory, or a stable memory (Non-Volatile Memory), such as a disk memory; the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0044] Those skilled in the art will understand that Figure 1 The device structure shown in the figure does not constitute a limitation of the device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0045] In a specific implementation, the device for automatically generating news channel information may be an ESP controller, or a vehicle-mounted computer, or other devices or terminals that realize the function of automatically generating news channel information, such as a central control unit or a remote server, which is not limited in this embodiment.

[0046] like Figure 1As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a program for automatically generating news channel information.

[0047] The device of the present invention calls the program for automatically generating news channel information stored in the memory 1005 through the processor 1001, and executes the operation steps of the method for automatically generating news channel information.

[0048] Based on the above hardware structure, an embodiment of the method for automatically generating news channel information of the present invention is proposed.

[0049] Reference Figure 2 , Figure 2 This is a flow chart of a first embodiment of a method for automatically generating news channel information according to the present invention.

[0050] Before executing the method for automatically generating news channel information of the present invention, a relational database is established. The ER diagram of the relational database is as follows: Figure 3 As shown:

[0051] Create a news event database: News_event; create a news picture library: News_picture; create a news audio library: News_audio. Here, take NetEase News as an example. Use the title of the hot news you want to generate a news channel to search for articles related to this news in NetEase News. The search result is the crawled target data. When the news data crawled from the source news title is stored in News_event, the label field is added. The content of the label field is the source news title, which is used to mark the content of the same channel.

[0052] In addition, news pages have several structures, which requires writing different xpath expressions to parse different pages, and determining which xpath expression to apply in the program. Use xpath to extract news text and splice it into a whole string. The references at the end of some news seriously affect the automatic summary effect, so they are removed using the rsplit method. The following is an example of returned data:

[0053] News list interface returns data

[0054]

[0055] Furthermore, in the first embodiment of the method for automatically generating news channel information of the present invention, the method for automatically generating news channel information comprises the following steps:

[0056] Step S1: determine the target news topic; obtain different source news titles corresponding to the target news topic respectively, crawl the corresponding content according to the time sequence of each source news title; determine whether the network address of each source news title already exists in the database;

[0057] It can be understood that the execution subject of this embodiment is the above-mentioned device for automatically generating news channel information, which first crawls related content in chronological order according to different source news titles, and determines whether the current URL (network address) already exists in the news event database;

[0058] Specifically, this embodiment uses an incremental crawler method to crawl the news content of the source news title:

[0059] Use the preset web crawler tool Scrapy. Scrapy will set up a set of url network addresses from startup to shutdown. The url network addresses crawled during Scrapy operation are placed in the network address set. Each time a request is made to a certain network address url to be processed, the network address to be processed is compared with the network addresses already in the network address set, and the duplicate network addresses that have been crawled are filtered out. In other words, Scrapy will maintain a set from startup to shutdown, and the urls crawled during operation are placed in the set. Each time a request is made to a certain url, it must be compared in the set set. When the crawler is started, the urls crawled in the database are stored in the set, and the crawled urls are filtered out.

[0060] Step S2: taking the source news title corresponding to the network address not existing in the database as the target source news title, determining the basic news information from the crawled content for each target source news title, selecting several sentences from the crawled content to generate a news summary of the target news topic, and storing the basic news information and the news summary in the database;

[0061] It is understandable that if S1 determines that the current URL does not exist in the database, it selects several sentences to generate a news summary, and stores the summary and basic news information in the database;

[0062] The summary of this embodiment adopts an extractive algorithm to calculate the weight of each sentence, select the most important sentences and splice them into a paragraph as the summary of this news. The number of summaries obtained in this step can be changed according to actual conditions.

[0063] In the specific implementation of this embodiment, before automatic summarization, it is first determined whether the total length of the article is less than 200 words. If the length of the article is less than 200 words, the entire article is used as a news summary. If the length of the article is greater than 200 words, 200 words are used as the maximum summary length, and 5 sentences with the highest weights are selected as summaries. If the total number of words is greater than the specified summary length, a sentence with a smaller weight is removed until the word count requirement is met. If only the last sentence is left and the length still exceeds the requirement, this sentence is used as the news summary.

[0064] It should be noted that in the TextRank algorithm of this embodiment, the news title is considered while calculating the sentence weight. The weight of the sentence is appropriately increased according to the relevance to the title, which can make the automatic summary result more reasonable. The weight of each sentence is calculated by the following formula:

[0065]

[0066] Where d is the damping coefficient, S0 is the number of words repeated in the source news title and the sentence, S i is the number of words in the ith sentence, In(V i ) and Out(V j ) are pointing to the node V i The node set and slave node V i The set of nodes pointed to by the outgoing edge, WS(V i ) represents node V i rank value, WS(V j ) represents node V j rank value, w ji Indicates V j To V i The weight is the similarity between sentences, W jk Indicates V j To V k The weight of S0 and S i The ratio of news headlines to sentence weights is used to realize the influence of news headlines on sentence weights.

[0067] The formula for examining sentence similarity is as follows:

[0068]

[0069] In the formula, S i and S j Represents the total number of words in two sentences, w kRepresents the words in the sentence. The numerator means the number of the same word that appears in both sentences at the same time, and the denominator is the sum of the logarithms of the number of words in the sentence. The design of the denominator can curb the advantage of longer sentences in similarity calculation. According to the above similarity formula, the similarity between any two nodes is calculated cyclically, and the edge connection with low similarity between the two nodes is removed according to the threshold. A node connection graph is constructed, and then the TextRank value is calculated. Finally, all TextRank values ​​are sorted, and the sentences corresponding to the 5 nodes with the highest TextRank values ​​are selected as summaries.

[0070] The automatically summarized data is stored in the abstract field of the corresponding news in the news event database News_event.

[0071] This step calculates the weight of each sentence in the article and selects sentences with higher weights to achieve automatic summarization of news text. By improving the TextRank algorithm, the automatic news summarization has a better effect.

[0072] Step S3, converting the text content of the news summary into voice information, and saving the language information;

[0073] Specifically, the text content of the summary is converted into speech through speech synthesis technology and saved in MP3 format. This step can use a mature prediction conversion tool, which will not be described in detail in this embodiment.

[0074] Step S4: crawl news event pictures related to the target news topic through the network address of each target source news title, and save the crawled news event pictures in the database; please refer to the attached Figure 4 ;

[0075] It is understandable that this step takes Baidu Pictures as an example. In this step, the title field can be used as a keyword to request Baidu Pictures. After obtaining the JSON data returned by Baidu Pictures, the picture URLs are parsed and downloaded and saved one by one. The process can be ended when 30 pictures are crawled.

[0076] It should be noted that although some image URLs are different, the image content may be the same, resulting in a high duplication rate. Solution: Cut off the part from "?sec=" to the end of the URL and cache it. When crawling the next image, compare them. If they are the same, skip them. In addition, there is a 20% probability that each URL will be skipped directly. This method can effectively reduce the duplication rate of image content;

[0077] Specifically, in this step, each time the current news event picture related to the target news topic is crawled, the preset field in the corresponding network address is removed and cached to obtain the network address to be compared; when the next news event picture related to the target news topic is crawled, the difference between the two network addresses to be compared is compared: if the comparison result is no difference, skip the current step and continue to crawl the next news event picture; if the comparison result is a difference, execute the step of saving the crawled news event picture in the database.

[0078] Assume that the final generated video size is 540x1080, so the picture needs to be proportionally enlarged or reduced to a width of 540 pixels and saved in News_picture. The path is the create_time field type serialized as a string and then concatenated with the root path.

[0079] Step S5: synthesizing a target news theme video according to the text content of the news summary, the voice information, and the news event picture;

[0080] Specifically, the text content of the news summary is used as the subtitle of the target news topic, the language information is used as the audio of the target news topic, and the news event picture is used as the background of the target news topic video.

[0081] In the specific implementation, the audio duration corresponding to the news is used as the duration of the entire video, and a size of 540x1080 pixels and an RGB value of (88,88,88) are used as the video background. The title is added to the appropriate position below the video, and the subtitles are split into sections and dynamically added to the appropriate position above the video. The image material is taken from the News_picture corresponding to create_time, and finally a special effect is randomly selected and added to the video.

[0082] During the video synthesis process, the switching of multiple images can be randomly processed by using up and down speed shifting, fading in and out, rotation, left and right uniform speed translation, cutting and other exit and entry methods.

[0083] It is understandable that most of the similar methods and systems currently require manual selection of templates before automatically generating videos. The method of this embodiment uses pictures crawled based on keywords in news titles as backgrounds, combined with random processing of transition animations, which can save the time of manually producing short videos, including the process of selecting templates, and more significantly reduces the manual labor in the news industry.

[0084] Step S6: inserting the generated target news theme video into the current news, marking the title of the target news theme and the corresponding time, and generating exclusive channel information related to the target news theme.

[0085] Specifically, the generated new video is inserted into the end of the current news in reverse chronological order, the title and time of the news at that time are marked, and an exclusive channel for the related news is generated.

[0086] This embodiment provides a method that can automatically integrate past news content summaries and can automatically broadcast them in audio and video. It can integrate the hot news currently being discussed, find relevant past information, summarize it into brief text, and process it into audio and video, which is more in line with the current fast-food way of obtaining information. Find past news based on current hot news, convert them into news summaries, and produce new audio and video. By implementing this method, the efficiency of obtaining information is improved, and the manual labor of the news industry is significantly reduced.

[0087] Further, based on the first embodiment of the method for automatically generating news channel information of the present invention, a second embodiment of the method for automatically generating news channel information of the present invention is proposed. In this embodiment, the step S10 specifically includes the following steps:

[0088] Before the step S5, it also includes: responding to the query request input by the user, searching for the corresponding news theme video from the database, and displaying the synthesized target news theme video; after receiving the user's selection operation on the video, executing the step S6.

[0089] It should be noted that the second method embodiment of the present invention belongs to a background management function for managing news data. After logging in, the administrator (i.e., the user) can crawl the news and news materials of the day. The obtained content can be fuzzy searched according to keywords, and the filtering conditions of the time period can be set to view the data of different time periods. The generated video and channel can be previewed, and the integrity of the video can be confirmed before publishing.

[0090] In this embodiment, when a user directly accesses the backend management through the URL, it will be intercepted and redirected to the login interface. Other pages can only be accessed normally after the user logs in.

[0091] All data is displayed in the center of the administrator page in the form of a table, with the table headers ID, title, abstract, create_time, label, and function. There is a query function above the table, where you can perform fuzzy queries using the title keyword, and you can also add time period filtering conditions. After checking "Channel", only source news data with channels will be displayed. After querying the data, you can click the "Publish" button to achieve one-click publishing of the news channel. In the function column, all data has a "Preview Video" button, which allows you to preview automatically generated videos. In data with consistent title and label fields, there is an additional "Preview Channel" button in the function, which allows you to preview the generated news channel.

[0092] Accordingly, the present invention also provides a first embodiment of a device for automatically generating news channel information, corresponding to the first embodiment of the method for automatically generating news channel information,

[0093] like Figure 5 As shown, the device comprises:

[0094] Crawler module: used to determine the target news topic; obtain different source news titles corresponding to the target news topic respectively, crawl the corresponding content according to the time sequence of each source news title; determine whether the network address of each source news title already exists in the database;

[0095] Automatic summary module: if it is determined that the target source news title corresponding to the network address does not exist in the database, then for each target source news title, basic news information is determined from the crawled content, and several sentences are selected from the crawled content to generate a news summary of the target news topic, and the basic news information and the news summary are stored in the database;

[0096] The synthetic video module converts the text content of the news summary into voice information and saves the language information; crawls news event pictures related to the target news topic through the network address of each target source news title, and saves the crawled news event pictures in the database; synthesizes the target news topic video according to the text content of the news summary, the voice information, and the news event pictures.

[0097] Furthermore, based on the first embodiment of the apparatus for automatically generating news channel information, a second embodiment of the apparatus of the present invention is proposed, corresponding to the second embodiment of the method:

[0098] In this embodiment, the device further includes:

[0099] The video publishing module is used to respond to the query request input by the user, search for the corresponding news theme video from the database, and display the synthesized target news theme video found; after receiving the user's selection operation on the video, the generated target news theme video is inserted into the current news, the title of the target news theme and the corresponding time are marked, and the exclusive channel information related to the target news theme is generated.

[0100] The device of this embodiment can integrate the hot news currently being discussed. Most importantly, it can find relevant past information, summarize it into concise text, and process it into audio and video, which is more in line with the current fast-food way of obtaining information.

[0101] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0102] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0103] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for automatically generating news channel information, characterized in that: The method comprises the following steps: Step S1: determine the target news topic; obtain different source news titles corresponding to the target news topic respectively, crawl the corresponding content according to the time sequence of each source news title; determine whether the network address of each source news title already exists in the database; Step S2: taking the source news title corresponding to the network address not existing in the database as the target source news title, determining the basic news information from the crawled content for each target source news title, selecting several sentences from the crawled content to generate a news summary of the target news topic, and storing the basic news information and the news summary in the database; Step S3: converting the text content of the news summary into voice information, and saving the language information; Step S4: crawling news event pictures related to the target news topic through the network address of each target source news title, and storing the crawled news event pictures in the database; Step S5: synthesizing a target news theme video according to the text content of the news summary, the voice information, and the news event picture; Step S6: inserting the generated target news theme video into the current news, marking the title of the target news theme and the corresponding time, and generating exclusive channel information related to the target news theme; The steps of generating the news summary of the target news topic in step S2 are: The summary uses an extractive algorithm to calculate the weight of each sentence, and selects several sentences according to the weight of each sentence to form a paragraph as the summary of this news. The weight of each sentence is calculated by the following formula: Where d is the damping coefficient, S0 is the number of words repeated in the source news title and the sentence, S i is the number of words in the ith sentence, In(V i ) and Out(V j ) are pointing to the node V i The node set and slave node V i The set of nodes pointed to by the departure edge, WS(Vi) represents the rank value of node Vi, WS(Vj) represents the rank value of node Vj, and w ji Indicates V j To V i The weight of is the similarity between sentences, Wjk represents the weight from Vj to Vk; through S0 and S i The ratio of news headlines to sentence weights is used to realize the influence of news headlines on sentence weights.

2. A method for automatically generating news channel information as claimed in claim 1, characterized in that: In step S1, the news content of the source news title is crawled by using an incremental crawler method, which specifically includes: A preset web crawler tool is used, which establishes a network address set from startup to shutdown, and the network addresses crawled during the operation of the preset web crawler tool are placed in the network address set. Each time a request is initiated to a network address to be processed, the network address to be processed is compared with the existing network addresses in the network address set, and duplicate network addresses that have been crawled are filtered out.

3. A method for automatically generating a news channel as claimed in claim 1, characterized in that: The step S4 further comprises: Each time a current news event picture related to the target news topic is crawled, the preset field in the corresponding network address is removed and cached to obtain the network address to be compared; When crawling the next news event image related to the target news topic, compare the difference between the two network addresses to be compared: If the comparison result is no difference, skip the current step and continue crawling the next news event picture; if the comparison result is a difference, execute the step of saving the crawled news event picture in the database.

4. A method for automatically generating a news channel as claimed in claim 1, characterized in that: In step S5, the text content of the news summary is used as the subtitle of the target news topic, the language information is used as the audio of the target news topic, and the news event picture is used as the background of the target news topic video.

5. A method for automatically generating a news channel as claimed in claim 1, characterized in that: Before step S5, the method further includes: In response to a query request input by a user, searching for a corresponding news theme video from the database, and displaying the found synthesized target news theme video; After receiving the user's selection operation on the video, the step S6 is executed.

6. A device for automatically generating news channel information, characterized in that: The device comprises: Crawler module: used to determine the target news topic; obtain different source news titles corresponding to the target news topic respectively, crawl the corresponding content according to the time sequence of each source news title; determine whether the network address of each source news title already exists in the database; Automatic summary module: if it is determined that the target source news title corresponding to the network address does not exist in the database, then for each target source news title, basic news information is determined from the crawled content, and several sentences are selected from the crawled content to generate a news summary of the target news topic, and the basic news information and the news summary are stored in the database; Synthetic video module: converting the text content of the news summary into voice information and saving the language information; crawling news event pictures related to the target news topic through the network address of each target source news title, and saving the crawled news event pictures in the database; synthesizing the target news topic video according to the text content of the news summary, the voice information, and the news event pictures; The automatic summary module is further used for: The summary uses an extractive algorithm to calculate the weight of each sentence, and selects several sentences according to the weight of each sentence to form a paragraph as the summary of this news. The weight of each sentence is calculated by the following formula: Where d is the damping coefficient, S0 is the number of words repeated in the source news title and the sentence, S i is the number of words in the ith sentence, In(V i ) and Out(V j ) are pointing to the node V i The node set and slave node V i The set of nodes pointed to by the departure edge, WS(Vi) represents the rank value of node Vi, WS(Vj) represents the rank value of node Vj, and w ji Indicates V j To V i The weight of is the similarity between sentences, Wjk represents the weight from Vj to Vk; through S0 and S i The ratio of news headlines to sentence weights is used to realize the influence of news headlines on sentence weights.

7. The device according to claim 6, characterized in that The device also includes: The video publishing module is used to respond to the query request input by the user, search for the corresponding news theme video from the database, and display the synthesized target news theme video found; after receiving the user's selection operation on the video, the generated target news theme video is inserted into the current news, the title of the target news theme and the corresponding time are marked, and the exclusive channel information related to the target news theme is generated.

8. A device for automatically generating news channel information, characterized in that: The device comprises: a memory, a processor and a program for automatically generating news channel information stored in the memory and executable on the processor, wherein the program is configured to implement the steps of the method for automatically generating news channel information as described in any one of claims 1 to 5.

9. A storage medium, characterized in that: The storage medium stores a program for automatically generating news channel information, and when the program is executed by the processor, the steps of the method for automatically generating news channel information according to any one of claims 1 to 5 are implemented.

Images