Analysis and directional propagation method for agriculture-related science popularization information

Abstract

The invention relates to an analysis and directional propagation method for agriculture-related science popularization information. The method comprises the following steps: judging whether a directional propagation demand is triggered or not; if so, determining to-be-propagated target agriculture-related science popularization information and a propagation object; determining other objects related to the propagation object according to the relation graph; and spreading the target agriculture-related science popularization information to the spreading object and other objects. According to the method provided by the invention, the spreading object of the agriculture-related science popularization information can be actively determined according to the relation graph, and the agriculture-related science popularization information is spread to the spreading object, so that active spreading of the agriculture-related science popularization information is realized, the spreading range is expanded, and the spreading effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method for analyzing and directional dissemination of agricultural science popularization information. Background Art

[0002] In recent years, the development of information technology has made great strides, and the construction of agricultural informatization is the top priority in building my country's modern agriculture.

[0003] Agricultural science popularization information can help agricultural workers improve their agricultural knowledge reserves and assist agricultural workers in making timely and accurate agricultural production decisions. Therefore, the dissemination of agricultural science popularization information is extremely important.

[0004] Traditionally, the dissemination of agricultural science popularization information is mainly passive, that is, the agricultural science popularization information is published on fixed websites (such as government propaganda websites). When users (such as agricultural workers) need to obtain agricultural science popularization information, they log in to the fixed website to query it themselves.

[0005] The traditional method relies on users' self-query to spread agricultural science information, but its dissemination effect is limited to existing query users, with a small dissemination range and poor dissemination effect. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In order to solve the above problems, the present invention provides a method for analyzing and directional dissemination of agricultural science popularization information.

[0008] (II) Technical solution

[0009] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0010] A method for analyzing and directional dissemination of agricultural science popularization information, the method comprising:

[0011] Determine whether the demand for targeted communication is triggered;

[0012] If triggered, the target agricultural science information to be disseminated and the dissemination targets are determined;

[0013] According to the relationship map, determine other objects related to the communication object;

[0014] Disseminate targeted agricultural science information to target audiences and other objects.

[0015] Optionally, determining whether the directional propagation demand is triggered includes:

[0016] If a query request sent by any user is received, it is determined that the directional propagation demand is triggered;

[0017] Determine the target agricultural science information to be disseminated and the dissemination targets, including:

[0018] Determine the agricultural science popularization information involved in the query request as the target agricultural science popularization information;

[0019] Identify any user as a target for dissemination.

[0020] Optionally, determining whether the directional propagation demand is triggered includes:

[0021] If it is confirmed that there is new agricultural science popularization information to be disseminated, it is determined that the demand for targeted dissemination is triggered;

[0022] Determine the target agricultural science information to be disseminated and the dissemination targets, including:

[0023] Determine the new agricultural science popularization information to be disseminated as the target agricultural science popularization information;

[0024] The users involved in the new agricultural science popularization information to be disseminated are determined as the dissemination targets.

[0025] Optionally, the method further comprises:

[0026] Get new agricultural science information in real time;

[0027] Determine, from among the currently stored agricultural science popularization information, the comparison agricultural science popularization information that belongs to the same category as the new agricultural science popularization information;

[0028] Extract the feature vector V1 of new agricultural science popularization information;

[0029] Determine V1 and compare the characteristic vector V of agricultural science information i The similarity between 1,i ; Wherein, i is the identifier of the agricultural science information to be compared;

[0030] like , it is determined that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated;

[0031] like , then segment the new agricultural science popularization information, determine whether each segmentation is an entity segmentation, and determine the entity type corresponding to each entity segmentation; determine the value of the new agricultural science popularization information based on the entity segmentation and the corresponding entity type; when , and when the value of the new agricultural science popularization information is not less than the value threshold, the new agricultural science popularization information is determined to be new agricultural science popularization information to be disseminated, and the new agricultural science popularization information is stored; otherwise, the new agricultural science popularization information is determined not to be new agricultural science popularization information to be disseminated.

[0032] Optionally, the value of new agricultural science popularization information is determined based on entity segmentation and corresponding entity types, including:

[0033] Determine the regional similarity SR of new agricultural science information based on V1;

[0034] Determine the publisher value ES1 of new agricultural science information;

[0035] Determine the value of new agricultural science information Among them, α1 is the regional weight, α2 is the entity segmentation quantity weight, α3 is the entity segmentation type weight, n1 is the total number of entity segmentations, n2 is the total number of segmentations of new agricultural science popularization information, n3 is the number of entity types corresponding to all entity segmentations, and n4 is the number of entity types corresponding to entity segmentations in the agricultural science popularization information with the largest similarity.

[0036] Optionally, if , after determining that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated, it also includes:

[0037] Determine the publisher value ES1 of the new agricultural science popularization information and the publisher value ES2 of the agricultural science popularization information with the greatest similarity;

[0038] Identify the semantics of new agricultural science popularization information and compare the semantics of agricultural science popularization information with the greatest similarity;

[0039] Determine the similarity SS between the semantics of the new agricultural science popularization information and the semantics of the agricultural science popularization information with the greatest similarity;

[0040] If SS>semantic threshold and ES1>ES2, the compared agricultural science information with the greatest similarity will be updated to the new agricultural science information.

[0041] Optionally, the relationship graph is composed of multiple nodes and edges;

[0042] Each point corresponds to a unique user;

[0043] If there is a relationship between two users, there is an edge between the points corresponding to the two users;

[0044] Each edge has an edge weight;

[0045] Among them, the edge weight of any edge e is the edge identifier, n5 is the number of agricultural science popularization information browsed by the users corresponding to the two points connected by edge e, n6 is the total number of agricultural science popularization information browsed by the users corresponding to one point connected by edge e, n7 is the total number of agricultural science popularization information browsed by the users corresponding to another point connected by edge e, n8 is the number of messages sent by the user corresponding to one point connected by edge e to the user corresponding to another point, n9 is the number of messages sent by the user corresponding to another point connected by edge e to the user corresponding to one point, n 10 The total number of messages sent by users corresponding to a point connected by edge e, n 11 is the total number of messages sent by the user corresponding to a point connected by edge e, and Δt is the time parameter of the users corresponding to the two points connected by edge e.

[0046] Optionally,

[0047] Among them, t1 is the effective duration of the user corresponding to a point connected by edge e, t2 is the effective duration of the user corresponding to another point connected by edge e, t3 is the registration duration of the user corresponding to a point connected by edge e, and t4 is the registration duration of the user corresponding to another point connected by edge e.

[0048] Optionally, other objects related to the propagation object are determined according to the relationship graph, including:

[0049] According to the propagation object, determine the target node in the relationship graph;

[0050] Take the target node as the selection node;

[0051] Determine other nodes that have edges with the selected node; wherein the other nodes are not selected nodes;

[0052] Determine the similarity SD between the user corresponding to each other node and the propagation object j ; where j is the identifier of other nodes, SD j is the similarity between the user corresponding to other node j and the propagation object;

[0053] Determine the stop value for each other node Among them, K is the number of edges between other nodes j and the target node, k is the identifier of the edge between other nodes j and the target node, and a k The maximum number of edges between the two points connected by edge k and the target node;

[0054] If the stop value of a certain other node is not less than the stop threshold, the other node is used as a new selection node, and the step of determining other nodes with edges to the selection node and subsequent steps are repeated; if all other nodes are stop nodes, the users corresponding to all selection nodes are other objects related to the propagation object;

[0055] Among them, the stop values ​​of the stop nodes are all smaller than the stop threshold, or the stop nodes have no other nodes with edges.

[0056] Optionally, according to the propagation object, a target node is determined in the relationship graph, including:

[0057] If there is a point corresponding to the propagation object in the relationship graph, then the point is determined to be the target node;

[0058] If the point corresponding to the propagation object does not exist in the relationship graph, the similarity between the nodes corresponding to the propagation object in the relationship graph is determined, and the point with the largest similarity is determined as the target node.

[0059] (III) Beneficial effects

[0060] The present invention relates to a method for analyzing and directional dissemination of agricultural science popularization information, the method comprising: determining whether a directional dissemination demand is triggered; if triggered, determining the target agricultural science popularization information to be disseminated and the dissemination object; determining other objects related to the dissemination object according to a relationship map; and disseminating the target agricultural science popularization information to the dissemination object and other objects. The method provided by the present invention can actively determine the dissemination object of agricultural science popularization information according to the relationship map, and disseminate the agricultural science popularization information to it, thereby realizing the active dissemination of agricultural science popularization information, expanding the dissemination scope, and improving the dissemination effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flowchart of a method for analyzing and disseminating agricultural science information provided in one embodiment of the present application;

[0062] Figure 2 A schematic diagram of a relationship graph provided for an embodiment of the present application. DETAILED DESCRIPTION

[0063] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0064] Traditional agricultural science popularization information is mainly disseminated in a passive way, that is, agricultural science popularization information is published on fixed websites (such as government propaganda websites). When users (such as agricultural workers) need to obtain agricultural science popularization information, they log in to the fixed website to query it themselves. The traditional method relies on users' self-query to achieve the dissemination of agricultural science popularization information. Its dissemination effect is only limited to existing query users, with a small dissemination range and poor dissemination effect.

[0065] To this end, the present invention relates to a method for analyzing and directional dissemination of agricultural science popularization information, the method comprising: determining whether the directional dissemination demand is triggered; if triggered, determining the target agricultural science popularization information to be disseminated and the dissemination object; determining other objects related to the dissemination object according to the relationship map; and disseminating the target agricultural science popularization information to the dissemination object and other objects. The method provided by the present invention can actively determine the dissemination object of agricultural science popularization information according to the relationship map, and disseminate the agricultural science popularization information to it, thereby realizing the active dissemination of agricultural science popularization information, expanding the dissemination scope, and improving the dissemination effect.

[0066] See also Figure 1 This embodiment provides a method for analyzing and directional dissemination of agricultural science popularization information. The implementation process of the method is as follows:

[0067] 101, determine whether the directional communication demand is triggered.

[0068] For example, if a query request sent by any user is received, it is determined that the directional communication demand is triggered. Alternatively, if it is confirmed that there is new agricultural science popularization information to be disseminated, it is determined that the directional communication demand is triggered.

[0069] That is, when a user (such as a farmer) makes a query, the demand for targeted dissemination is triggered; or, when there is new agricultural science information to be disseminated, the demand for targeted dissemination is triggered.

[0070] 102. If triggered, determine the target agricultural science information to be disseminated and the dissemination object.

[0071] Different triggering methods correspond to different determination schemes.

[0072] For example, if in step 101, it is determined that the directional communication demand is triggered when a query request sent by any user is received, then in step 102, the agricultural science popularization information involved in the query request is determined as the target agricultural science popularization information, and any user is determined as the communication object.

[0073] For another example, if in step 101, it is determined that the demand for targeted dissemination is triggered when it is confirmed that there is new agricultural science popularization information to be disseminated, then in step 102, the new agricultural science popularization information to be disseminated is determined as the target agricultural science popularization information, and the users involved in the new agricultural science popularization information to be disseminated are determined as the dissemination objects.

[0074] Among them, the users involved in the new agricultural science popularization information to be disseminated can be: users involved in the content of the new agricultural science popularization information to be disseminated, such as if the new agricultural science popularization information to be disseminated is XXXX of a wheat grower, then the users involved in the new agricultural science popularization information to be disseminated are wheat growers. For another example, if the new agricultural science popularization information to be disseminated is XXXX of a corn, then the users involved in the new agricultural science popularization information to be disseminated are corn-related users (such as growers, transport personnel, etc.). The users involved in the new agricultural science popularization information to be disseminated can also be: users involved in the surrounding areas of the new agricultural science popularization information to be disseminated, such as registered users of the publishing platform of the new agricultural science popularization information to be disseminated, users who follow the publisher of the new agricultural science popularization information to be disseminated, etc.

[0075] In addition, the new agricultural science popularization information to be disseminated is not all new agricultural science popularization information, but agricultural science popularization information that is valuable to be disseminated. For example, the new agricultural science popularization information to be disseminated is determined through the following process:

[0076] 1. Obtain new agricultural science information in real time.

[0077] For example, whenever a user uploads a new piece of popular science information related to agriculture, the new popular science information related to agriculture will be obtained.

[0078] 2. Among the currently stored agricultural science popularization information, determine the comparison agricultural science popularization information that belongs to the same category as the new agricultural science popularization information.

[0079] When new agricultural science information is uploaded, it will be asked to determine its category, such as insect control, science, etc. This step will find the agricultural science information of the same category for comparison.

[0080] In addition, the currently stored agricultural science popularization information is valuable agricultural science popularization information, not all agricultural science popularization information. In other words, the method provided in this embodiment will maintain a valuable agricultural science popularization information database in real time. As long as a new valuable agricultural science popularization information is disseminated, it will be put into the database. This step will find the same category of agricultural science popularization information from the database to compare agricultural science popularization information.

[0081] 3. Extract the feature vector V1 of new agricultural science information.

[0082] 4. Compare V1 with the characteristic vector V of agricultural science information i The similarity between 1,i .

[0083] Among them, i is the identifier for comparing agricultural science popularization information.

[0084] The similarity is determined using an existing calculation scheme.

[0085] 5. According to SIM 1,i Identify new agricultural science information to be disseminated.

[0086] 1) If , it means that the new agricultural science popularization information is very similar to the currently stored agricultural science popularization information. Since similar agricultural science popularization information has been stored and disseminated in the database, the agricultural science popularization information will no longer be disseminated. Therefore, it is determined that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated.

[0087] In addition, after determining that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated, since it is very similar to the stored agricultural science popularization information, it will also be determined whether it replaces the stored agricultural science popularization information to ensure that the stored agricultural science popularization information is the latest information. The judgment process is:

[0088] (1) Determine the publisher value ES1 of the new agricultural science popularization information and the publisher value ES2 of the agricultural science popularization information with the greatest similarity.

[0089] This value can be realized through existing methods, such as whether the publisher has been authenticated, the amount of information the publisher has published, etc. The publisher value reflects the dissemination value of new agricultural science popularization information. If the publisher value is smaller, the dissemination value of the new agricultural science popularization information is lower (such as the lower the authenticity, the less valuable information of the new agricultural science popularization information, etc.), and the new agricultural science popularization information is less likely to be the agricultural science popularization information to be disseminated.

[0090] (2) Identify the semantics of new agricultural science popularization information and compare the semantics of agricultural science popularization information with the greatest similarity.

[0091] Existing semantic recognition schemes can be used to identify the semantics of agricultural science information.

[0092] (3) Determine the similarity SS between the semantics of the new agricultural science popularization information and the semantics of the agricultural science popularization information with the greatest similarity.

[0093] The similarity SS may be calculated using an existing similarity calculation scheme.

[0094] (4) If SS>semantic threshold, and ES1>ES2, it means that the semantics of the new agricultural science popularization information is very similar to that of the stored comparison agricultural science popularization information, and the publisher of the new agricultural science popularization information is more valuable than the publisher of the stored comparison agricultural science popularization information. In this way, the comparison agricultural science popularization information with the greatest similarity will be updated to the new agricultural science popularization information.

[0095] The semantic threshold is a preset value.

[0096] 2) If , it means that there is no similar agricultural science popularization information in the current database, then the new agricultural science popularization information may be valuable for dissemination, and the following judgment will be made at this time:

[0097] 201, segment the new agricultural science popularization information, determine whether each segmentation is an entity segmentation, and determine the entity type corresponding to each entity segmentation.

[0098] Among them, entity participles are participles that can reflect actual meanings, such as nouns, etc. In specific implementation, a whitelist of parts of speech can be set. As long as the part of speech of a participle is in the whitelist, it is an entity participle, and the corresponding part of speech and entity type.

[0099] 202, determine the value of new agricultural science information based on entity segmentation and corresponding entity types.

[0100] (1) Determine the regional similarity SR of new agricultural science information based on V1.

[0101] This step determines the regional similarity SR through the existing scheme. The regional similarity SR reflects whether the new agricultural science popularization information is relevant to the current region. The more relevant it is, the more likely it is to be the agricultural science popularization information to be disseminated.

[0102] (2) Determine the publisher value ES1 of new agricultural science information.

[0103] This value can be realized through existing methods, such as whether the publisher has been authenticated, the amount of information the publisher has published, etc. The publisher value reflects the dissemination value of new agricultural science popularization information. If the publisher value is smaller, the dissemination value of the new agricultural science popularization information is lower (such as the lower the authenticity, the less valuable information of the new agricultural science popularization information, etc.), and the new agricultural science popularization information is less likely to be the agricultural science popularization information to be disseminated.

[0104] (3) Determine the value of new agricultural science information

[0105] Among them, α1 is the regional weight, α2 is the entity segmentation quantity weight, α3 is the entity segmentation type weight, n1 is the total number of entity segmentations, n2 is the total number of segmentations of new agricultural science popularization information, n3 is the number of entity types corresponding to all entity segmentations, and n4 is the number of entity types corresponding to entity segmentations in the agricultural science popularization information with the largest similarity.

[0106] α1, α2, α3 can be preset.

[0107] The larger the SR, the closer the regions are. The larger the SR, the greater the dissemination value of new agricultural science information. It reflects the proportion of entity segmentations of new agricultural science popularization information. The larger the value is, the more entity segmentations are included in the new agricultural science popularization information, which means that the dissemination value of the new agricultural science popularization information is greater. It is the similarity between the entity segmentation of the new agricultural science popularization information and the entity segmentation of the agricultural science popularization information that may be spread (i.e. the agricultural science popularization information with the largest similarity). The larger the value, the more similar it is and the more likely it is to be spread. Therefore, ET reflects the value of the new agricultural science popularization information.

[0108] 203, when , and the value of the new agricultural science popularization information is not less than the value threshold, It represents that the similarity between the new agricultural science popularization information and the comparison agricultural science popularization information fluctuates greatly, that is, the similarity degree of the comparison agricultural science popularization information may be extremely different, indicating that it is not similar to the agricultural science popularization information that has been disseminated (that is, the agricultural science popularization information stored in the database), and the fluctuation is relatively large, but the value of the new agricultural science popularization information is relatively high, that is, similar agricultural science popularization information has not been disseminated at present, but the value of the new agricultural science popularization information is large, then it needs to be disseminated, so the new agricultural science popularization information is determined as the new agricultural science popularization information to be disseminated, and the new agricultural science popularization information is stored (that is, the new agricultural science popularization information is stored in the database).

[0109] Otherwise, it is determined that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated.

[0110] 103. According to the relationship graph, determine other objects related to the communication object.

[0111] Among them, the relationship map is also pre-established and maintained in real time.

[0112] like Figure 2 As shown in the figure, the relationship graph consists of multiple points and edges. Each point corresponds to a unique user. If there is a relationship between two users, there is an edge between the points corresponding to the two users.

[0113] Each edge has an edge weight.

[0114] Among them, the edge weight of any edge

[0115] e is the edge identifier, n5 is the number of agricultural science popularization information browsed by the users corresponding to the two points connected by edge e, n6 is the total number of agricultural science popularization information browsed by the users corresponding to one point connected by edge e, n7 is the total number of agricultural science popularization information browsed by the users corresponding to another point connected by edge e, n8 is the number of messages sent by the user corresponding to one point connected by edge e to the user corresponding to another point, n9 is the number of messages sent by the user corresponding to another point connected by edge e to the user corresponding to one point, n 10The total number of messages sent by users corresponding to a point connected by edge e, n 11 is the total number of messages sent by the user corresponding to a point connected by edge e, and Δt is the time parameter of the users corresponding to the two points connected by edge e.

[0116]

[0117] Among them, t1 is the effective duration of the user corresponding to a point connected by edge e, t2 is the effective duration of the user corresponding to another point connected by edge e, t3 is the registration duration of the user corresponding to a point connected by edge e, and t4 is the registration duration of the user corresponding to another point connected by edge e.

[0118] For example, for the edge between point 3 and point 10, n5 is the number of agricultural science popularization information browsed by the users corresponding to point 3 and point 10 connected by the edge, n6 is the total number of agricultural science popularization information browsed by the users corresponding to point 3, n7 is the total number of agricultural science popularization information browsed by the users corresponding to point 10, n8 is the number of messages sent by the users corresponding to point 3 to the users corresponding to point 10, n9 is the number of messages sent by the users corresponding to point 10 to the users corresponding to point 3, and n 10 is the total number of messages sent by the user corresponding to point 3, n 11 is the total number of messages sent by the user corresponding to point 10, and Δt is the time parameter of the users corresponding to the two points connected by edge e.

[0119] It represents the similarity between the users corresponding to point 3 and point 10. It represents the communication degree of the users corresponding to point 3 and point 10, and Δt represents the activity degree of the users corresponding to point 3 and point 10. Edge weight w e It represents the communication value between users corresponding to point 3 and point 10.

[0120] The implementation process of step 103 is:

[0121] 103-1, determine the target node in the relationship graph according to the propagation object.

[0122] For example, if the point corresponding to the propagation object exists in the relationship graph, the point is determined as the target node. If the point corresponding to the propagation object does not exist in the relationship graph, the similarity between the corresponding user of each node in the relationship graph and the propagation object is determined, and the point with the largest similarity is determined as the target node.

[0123] 103-2, the target node is used as the selection node.

[0124] 103-3, determine other nodes that have edges with the selected node.

[0125] Among them, other nodes are non-selected nodes.

[0126] 103-4, determine the similarity SD between the user corresponding to each other node and the propagation object j .

[0127] Among them, j is the identifier of other nodes, SD j is the similarity between the user corresponding to other node j and the propagation object.

[0128] 103-5, determine the stop value of each other node

[0129] Among them, K is the number of edges between other nodes j and the target node, k is the identifier of the edge between other nodes j and the target node, and a k The maximum number of edges between the two points connected by edge k and the target node.

[0130] 103-6, if the stop value of a certain other node is not less than the stop threshold, the other node is used as a new selection node, and the step of determining other nodes with edges to the selection node and subsequent steps are repeated (i.e., steps 103-3 to 103-6 are repeated). If all other nodes are stop nodes, the users corresponding to all selection nodes are other objects related to the propagation object.

[0131] Among them, the stop values ​​of the stop nodes are all smaller than the stop threshold, or the stop nodes have no other nodes with edges.

[0132] For example, in step 103-1, the target node is determined to be point 4. In step 103-2, point 4 is selected as a selection node. In step 103-3, other nodes (i.e., point 1 and point 9) that have edges with point 4 are determined. In step 103-4, the similarity SD1 between the user corresponding to point 1 and the propagation object and the similarity SD9 between the user corresponding to point 9 and the propagation object are determined. In step 103-5, the stop value ST1 of point 1 and the stop value ST9 of point 9 are determined. If it is determined in step 106-6 that ST1 is not less than the stop threshold and ST9 is not less than the stop threshold, then both point 1 and point 9 are used as new selection nodes, and step 103-3 is repeated to determine other nodes that have edges with point 9. Since point 9 has no other nodes, point 9 is a stop node, and processing is stopped. Step 103-3 is repeated to determine other nodes (i.e., point 2 and point 5) that have edges with point 1. In step 103-4, the similarity SD2 between the user corresponding to point 2 and the propagation object, and the similarity SD5 between the user corresponding to point 5 and the propagation object are determined. In step 103-5, the stop value ST2 of point 2 and the stop value ST5 of point 5 are determined. If it is determined in step 106-6 that ST2 is not less than the stop threshold, but ST5 is less than the stop threshold, then point 5 is a stop node and processing stops (but point 5 will not be used as a new selection node). Point 2 is used as a new selection node, and step 103-3 is repeated to determine other nodes that have edges with point 2. Since point 2 has no other nodes, point 2 is a stop node and processing stops.

[0133] The users corresponding to all selected nodes (ie, point 4, point 9, point 1, and point 2) are all other objects related to the propagation object.

[0134] 104. Disseminate targeted agricultural science information to target audiences and other targets.

[0135] This step can adopt existing communication solutions, such as pushing target agricultural science information to them through messages.

[0136] This embodiment provides a method for analyzing and directional dissemination of agricultural science popularization information, which determines whether the directional dissemination demand is triggered; if triggered, determines the target agricultural science popularization information to be disseminated and the dissemination object; determines other objects related to the dissemination object based on the relationship map; and disseminates the target agricultural science popularization information to the dissemination object and other objects. The method provided in this embodiment can actively determine the dissemination object of agricultural science popularization information based on the relationship map, and disseminate agricultural science popularization information to it, thereby realizing the active dissemination of agricultural science popularization information, expanding the dissemination scope, and improving the dissemination effect.

[0137] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0138] It should also be noted that the exemplary embodiments mentioned in the present invention describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in a different order from the embodiments, or several steps can be performed simultaneously.

[0139] Finally, it should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing and directional dissemination of agricultural science popularization information, characterized in that: The method comprises: Determine whether the demand for targeted communication is triggered; If triggered, the target agricultural science information to be disseminated and the dissemination targets are determined; Determine other objects related to the communication object according to the relationship map; Disseminate the target agricultural science information to the dissemination objects and other objects.

2. The method according to claim 1, characterized in that The determining whether the directional propagation demand is triggered includes: If a query request sent by any user is received, it is determined that the directional propagation demand is triggered; Determining the target agricultural science popularization information to be disseminated and the dissemination objects includes: Determining the agricultural science popularization information involved in the query request as target agricultural science popularization information; Determine any one of the users as a propagation target.

3. The method according to claim 1, characterized in that The determining whether the directional propagation demand is triggered includes: If it is confirmed that there is new agricultural science popularization information to be disseminated, it is determined that the demand for targeted dissemination is triggered; Determining the target agricultural science popularization information to be disseminated and the dissemination objects includes: Determining the new agricultural science popularization information to be disseminated as target agricultural science popularization information; The users involved in the new agricultural science popularization information to be disseminated are determined as dissemination targets.

4. The method according to claim 3, characterized in that: The method further comprises: Get new agricultural science information in real time; Determine, from among the currently stored agricultural science popularization information, the comparison agricultural science popularization information that belongs to the same category as the new agricultural science popularization information; Extract the feature vector V1 of new agricultural science popularization information; Determine V1 and compare the characteristic vector V of agricultural science information i The similarity between 1,i ; Wherein, i is the identifier of the agricultural science information to be compared; like It is determined that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated; like Then, the new agricultural science popularization information is segmented to determine whether each segmentation is an entity segmentation, and the entity type corresponding to each entity segmentation is determined; the value of the new agricultural science popularization information is determined according to the entity segmentation and the corresponding entity type; when And when the value of the new agricultural science popularization information is not less than the value threshold, the new agricultural science popularization information is determined to be new agricultural science popularization information to be disseminated, and the new agricultural science popularization information is stored; otherwise, it is determined that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated.

5. The method according to claim 4, characterized in that The determining the value of the new agricultural science popularization information according to the entity segmentation and the corresponding entity type includes: Determine the regional similarity SR of the new agricultural science popularization information according to V1; Determine the publisher value ES1 of the new agricultural science information; Determine the value of the new agricultural science information Among them, α1 is the regional weight, α2 is the entity segmentation quantity weight, α3 is the entity segmentation type weight, n1 is the total number of entity segmentations, n2 is the total number of segmentations of the new agricultural science popularization information, n3 is the number of entity types corresponding to all entity segmentations, and n4 is the number of entity types corresponding to the entity segmentations in the agricultural science popularization information with the largest similarity.

6. The method according to claim 4, characterized in that like After determining that the new agricultural science popularization information is not new agricultural science popularization information to be disseminated, the method further includes: Determine the publisher value ES1 of the new agricultural science popularization information and the publisher value ES2 of the compared agricultural science popularization information with the greatest similarity; Identify the semantics of the new agricultural science popularization information and the semantics of the agricultural science popularization information with the greatest similarity; Determine the similarity SS between the semantics of the new agricultural science popularization information and the semantics of the agricultural science popularization information with the greatest similarity; If SS>semantic threshold, and ES1>ES2, the compared agricultural science information with the greatest similarity is updated to the new agricultural science information.

7. The method according to claim 1, characterized in that The relationship graph is composed of multiple points and edges; Each point corresponds to a unique user; If there is a relationship between two users, there is an edge between the points corresponding to the two users; Each edge has an edge weight; Among them, the edge weight of any edge e is the edge identifier, n5 is the number of agricultural science popularization information browsed by the users corresponding to the two points connected by edge e, n6 is the total number of agricultural science popularization information browsed by the users corresponding to one point connected by edge e, n7 is the total number of agricultural science popularization information browsed by the users corresponding to another point connected by edge e, n8 is the number of messages sent by the user corresponding to one point connected by edge e to the user corresponding to another point, n9 is the number of messages sent by the user corresponding to another point connected by edge e to the user corresponding to one point, n 10 The total number of messages sent by users corresponding to a point connected by edge e, n 11 is the total number of messages sent by the user corresponding to a point connected by edge e, and Δt is the time parameter of the users corresponding to the two points connected by edge e.

8. The method according to claim 7, characterized in that Said Among them, t1 is the effective duration of the user corresponding to a point connected by edge e, t2 is the effective duration of the user corresponding to another point connected by edge e, t3 is the registration duration of the user corresponding to a point connected by edge e, and t4 is the registration duration of the user corresponding to another point connected by edge e.

9. The method according to claim 7, characterized in that: Determining other objects related to the propagation object according to the relationship graph includes: According to the propagation object, determining a target node in the relationship graph; Taking the target node as a selection node; Determine other nodes that have edges with the selected node; wherein the other nodes are not selected nodes; Determine the similarity SD between the user corresponding to each other node and the propagation object j ; where j is the identifier of other nodes, SD j is the similarity between the user corresponding to other node j and the propagation object; Determine the stop value for each other node Among them, K is the number of edges between other nodes j and the target node, k is the identifier of the edge between other nodes j and the target node, and a k The maximum number of edges between the two points connected by edge k and the target node; If the stop value of a certain other node is not less than the stop threshold, the other node is used as a new selection node, and the step of determining other nodes with edges to the selection node and subsequent steps are repeated; if all other nodes are stop nodes, the users corresponding to all selection nodes are other objects related to the propagation object; Among them, the stop values ​​of the stop nodes are all smaller than the stop threshold, or the stop nodes have no other nodes with edges.

10. The method according to claim 9, characterized in that The step of determining a target node in the relationship graph according to the propagation object includes: If a point corresponding to the propagation object exists in the relationship graph, the point is determined to be a target node; If the point corresponding to the propagation object does not exist in the relationship graph, the similarity between the points corresponding to each node in the relationship graph and the propagation object is determined, and the point with the largest similarity is determined as the target node.

Images