CN113761214A

CN113761214A - Information flow extraction method, device and equipment

Info

Publication number: CN113761214A
Application number: CN202010503615.1A
Authority: CN
Inventors: S·德斯潘德; 庞卡·库玛; 希夫尚卡尔·乌玛尚卡尔; 马库斯·汉斯
Original assignee: Smart Bud Information Technology Suzhou Co ltd
Current assignee: Smart Bud Information Technology Suzhou Co ltd
Priority date: 2020-06-05
Filing date: 2020-06-05
Publication date: 2021-12-07
Also published as: WO2021244657A1

Abstract

The embodiment of the specification discloses an information flow extraction method, a device and equipment, wherein the method comprises the steps of obtaining target entity information; extracting at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information; and associating the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information. By utilizing the embodiments of the specification, the accuracy and comprehensiveness of information extraction can be greatly improved.

Description

Information flow extraction method, device and equipment

Technical Field

The present disclosure relates to the field of computer data processing technologies, and in particular, to a method, an apparatus, and a device for extracting an information stream.

Background

The information flow extraction is a common means for a user to perform technical query, infringement analysis and the like, such as querying a patent layout of a competitor, whether a certain novel material processing technology exists in the prior art or not, and the like. However, generally, information may have a plurality of different description forms in the same data source or different data sources, and even a large amount of incomplete and non-descriptive information exists, so that the searched information has the problems of inaccuracy and incompleteness. And the searched information amount is possibly huge and irrelevant or the noise information is more, so that the user needs to expend more energy to screen and comb when checking, and the experience sense is poor.

At present, there are some methods for extracting and searching information, such as information query using keywords. However, since many specific information description formats in more data sources are complicated and varied, and there are problems of inaccuracy and incompleteness, it is difficult to accurately and comprehensively extract the information required by the user.

Disclosure of Invention

An object of the embodiments of the present disclosure is to provide an information flow extraction method, apparatus, and device, which can greatly improve accuracy and comprehensiveness of information query.

The present specification provides an information stream extraction method, apparatus and device, which are implemented in the following manner:

an information flow presentation method, comprising:

the method comprises the steps that a client sends an information flow obtaining request to a server, wherein the information flow obtaining request comprises input information obtained by the client;

the server receives the information flow acquisition request, and extracts one or more entity information from the input information as target entity information;

the server extracts at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information;

and the server associates the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information, and sends the information stream to the client for displaying on the client.

On the other hand, an embodiment of the present specification further provides an information flow extraction method, including:

acquiring target entity information;

extracting at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information;

and associating the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information.

In another embodiment of the method provided in this specification, the associating the target entity information and the associated entity information by using the association relationship information includes:

acquiring at least one associated entity information extracted from a data source and associated relationship information between the target entity information and the corresponding associated entity information, and associating the target entity information with the corresponding associated entity information by using the associated relationship information between the target entity information and the corresponding associated entity information to obtain a sub information stream of the target entity information;

and linking sub information streams obtained from the same or different data sources by taking the target entity information as reference information to obtain an information stream corresponding to the target entity information.

In another embodiment of the method provided in this specification, when the association relationship information includes peer description information, the association entity information corresponding to the peer description information is supplemented to the target entity information, and the target entity information is updated; the collocated entity description information comprises description information describing other expression forms of associated entity information as target entity information;

correspondingly, at least one piece of associated entity information corresponding to the updated target entity information and associated relation information between the target entity information and the associated entity information are extracted from the data source.

In another embodiment of the method provided in this specification, the method further comprises:

extracting information streams by taking each associated entity information extracted from the data source as target entity information to obtain information streams corresponding to a plurality of target entity information;

and linking information streams corresponding to the target entity information by taking the corresponding target entity information as reference information to obtain an information map.

In another embodiment of the method provided in this specification, the extracting, from a data source, at least one associated entity information corresponding to the target entity information and association relationship information between the target entity information and the associated entity information includes:

retrieving a data source where the target entity information is located;

locating context information of the target entity information in a data source;

and when the context information comprises a material process entity, a material name entity and/or a material application entity, extracting associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information according to a material process information association mode.

extracting the target entity information and the trade name of the related entity information from a data source;

and extracting a manufacturer or a supplier corresponding to the target entity information or the associated entity information according to the trade name, and associating the manufacturer or the supplier with the target entity information or the associated entity information.

extracting the target entity information and the parameter entity information of the associated entity information from a data source, wherein the parameter entity information at least comprises entity information corresponding to one of a material structure type entity, a process method entity, a material attribute entity, a unit entity or a measurement entity;

and associating the extracted parameter entity information with corresponding target entity information and associated entity information.

In another embodiment of the method provided in this specification, the method further comprises: and associating the sub information flow of the target entity information with a corresponding data source.

taking each entity information in the information flow as an information node interacting with a user, and carrying out visualization processing on the information flow by using a visualization method;

sending the processed information flow to a client so that a user can check the information flow and trigger the information node through the client;

and feeding back other information related to the information node to the client based on the triggering operation of the user on the information node.

On the other hand, an embodiment of the present specification further provides an information flow extraction apparatus, including:

the first acquisition module is used for acquiring target entity information;

a first extraction module, configured to extract, from a data source, at least one piece of associated entity information corresponding to the target entity information and associated relationship information between the target entity information and the associated entity information, where the associated relationship information includes information describing an association direction and an association type between entity information;

and the first association module is used for associating the target entity information with the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information.

In another embodiment of the apparatus provided in this specification, the first associating module includes:

the first association unit is used for acquiring at least one associated entity information extracted from a data source and association relationship information between the target entity information and the corresponding associated entity information, and associating the target entity information with the corresponding associated entity information by using the association relationship information between the target entity information and the corresponding associated entity information to acquire a sub information stream of the target entity information;

and the second association unit is used for linking sub information streams obtained from the same or different data sources by taking the target entity information as reference information to obtain an information stream corresponding to the target entity information.

In another embodiment of the apparatus provided in this specification, when the association relationship information includes a peer description information, the apparatus further includes:

an updating module, configured to supplement, to the target entity information, associated entity information corresponding to the peer description information, and update the target entity information; the collocated entity description information comprises description information describing other expression forms of associated entity information as target entity information;

the first extraction module is further configured to extract at least one associated entity information corresponding to the updated target entity information and association relationship information between the target entity information and the associated entity information from a data source.

In another embodiment of the apparatus provided in this specification, the apparatus further comprises:

the second extraction module is used for extracting information streams by taking each piece of associated entity information extracted from the data source as target entity information to obtain information streams corresponding to a plurality of pieces of target entity information;

and the second correlation module is used for linking the information streams corresponding to the target entity information by taking the corresponding target entity information as reference information to obtain an information map.

In another embodiment of the apparatus provided in this specification, the first extraction module includes:

the retrieval unit is used for retrieving the data source where the target entity information is located;

the positioning unit is used for positioning the context information of the target entity information in the data source;

and the extracting unit is used for extracting associated entity information corresponding to the target entity information and association relation information between the target entity information and the associated entity information according to a material process information association mode when the context information comprises a material process entity, a material name entity and/or a material application entity.

the third extraction module is used for extracting the trade names of the target entity information and the associated entity information from a data source;

and the third association module is used for extracting a manufacturer or a supplier corresponding to the target entity information or the associated entity information according to the trade name and associating the manufacturer or the supplier with the target entity information or the associated entity information.

a fourth extraction module, configured to extract, from a data source, parameter entity information of the target entity information and the associated entity information, where the parameter entity information at least includes entity information corresponding to one of a material structure type entity, a process method entity, a material attribute entity, a unit entity, and a measurement entity;

and the fourth correlation module is used for correlating the extracted parameter entity information with corresponding target entity information and correlation entity information.

In another embodiment of the apparatus provided in this specification, the first associating module further includes:

and the third association unit is used for associating the sub information flow of the target entity information with the corresponding data source.

the visualization processing module is used for taking each entity information in the information flow as an information node interacting with a user and carrying out visualization processing on the information flow by using a visualization method;

the first sending module is used for sending the processed information flow to a client so that a user can check the information flow and trigger the information node through the client;

and the second sending module is used for feeding back other information related to the information node to the client based on the triggering operation of the user on the information node.

In another aspect, an embodiment of the present specification further provides an information flow extraction apparatus, where the apparatus includes a processor and a memory for storing processor-executable instructions, and the instructions, when executed by the processor, implement one or more of the steps of the method described above.

On the other hand, an embodiment of the present specification further provides an information flow presentation method, which is applied to a server, and includes:

receiving an information flow acquisition request sent by a client, wherein the information flow acquisition request comprises input information acquired by the client;

extracting one or more entity information from the input information as target entity information;

associating the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information;

and sending the information flow to the client side for displaying on the client side.

In another aspect, an embodiment of the present specification further provides a server, including:

the first receiving module is used for receiving an information flow obtaining request sent by a client, wherein the information flow obtaining request comprises input information obtained by the client;

a fifth extraction module, configured to extract one or more pieces of entity information from the input information, as target entity information;

a sixth extraction module, configured to extract, from a data source, at least one piece of associated entity information corresponding to the target entity information and associated relationship information between the target entity information and the associated entity information, where the associated relationship information includes information describing an association direction and an association type between entity information;

a fifth association module, configured to associate the target entity information and the associated entity information by using the association relationship information, and obtain an information stream corresponding to the target entity information;

and the third sending module is used for sending the information flow to the client so as to display on the client.

On the other hand, an embodiment of the present specification further provides an information flow presentation method, which is applied to a client, and includes:

sending an information flow acquisition request to a server, wherein the information flow acquisition request comprises input information acquired by a client; enabling the server to receive the information flow obtaining request and extract one or more entity information from the input information to serve as target entity information; extracting at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information; associating the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information, and sending the information stream to the client;

and receiving the information flow sent by the server and displaying the information flow.

In another aspect, an embodiment of the present specification further provides a client, including:

a fourth sending module, configured to send an information flow obtaining request to a server, where the information flow obtaining request includes input information obtained by a client; enabling the server to receive the information flow obtaining request and extract one or more entity information from the input information to serve as target entity information; extracting at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information; associating the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information, and sending the information stream to the client;

the second receiving module is used for receiving the information flow sent by the server;

and the first display module is used for displaying the information flow.

On the other hand, an embodiment of the present specification further provides an information retrieval method, which is applied to a server, and includes:

receiving a retrieval request sent by a client, wherein the retrieval request comprises input information acquired by the client;

extracting entity information from the input information, wherein the entity information comprises an entity type and an entity value;

determining a data source where the entity information is located according to associated entity information of the entity information in the data source and associated relation information between the entity information and the associated entity information, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information;

and sending the retrieval result to the client to display the client.

the third receiving module is used for receiving a retrieval request sent by the client, wherein the retrieval request comprises input information acquired by the client;

a seventh extraction module, configured to extract entity information from the input information, where the entity information includes an entity type and an entity value;

the data source determining module is used for determining a data source where the entity information is located according to associated entity information of the entity information in the data source and associated relation information between the entity information and the associated entity information, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information;

and the fifth sending module is used for sending the retrieval result to the client so as to display the client.

On the other hand, an embodiment of the present specification further provides an information retrieval method, which is applied to a client, and includes:

sending a retrieval request to a server, wherein the retrieval request comprises input information acquired by a client; so that the server extracts entity information from the input information, wherein the entity information comprises an entity type and an entity value; determining a data source where the entity information is located according to associated entity information of the entity information in the data source and associated relation information between the entity information and the associated entity information, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information; sending the retrieval result to the client;

and receiving and displaying the retrieval result sent by the server.

In another embodiment of the method provided in this specification, the information input interface displayed by the client includes an information input area, a first selection list and/or a second selection list; the information input area is used for inputting information; the first selection list and the second selection list are used for information selection, and the first selection list comprises category information of material application or material structures; the second selection list comprises category information of entity types;

correspondingly, the client side obtains input information based on the information input area, the first selection list and/or the second selection list.

In another embodiment of the method provided in the present specification, the first selection list and/or the second selection list show information to be selected in a format including interactive visualization; the interactive visualization format shows entity types or material structures and category information of material applications in a visualization form, and determines a list selection format of corresponding categories and subcategories by receiving trigger operations on the category information.

a sixth sending module, configured to send a retrieval request to the server, where the retrieval request includes input information obtained by the client; so that the server extracts entity information from the input information, wherein the entity information comprises an entity type and an entity value; determining a data source where the entity information is located according to associated entity information of the entity information in the data source and associated relation information between the entity information and the associated entity information, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information; sending the retrieval result to the client;

the fourth receiving module is used for receiving the retrieval result sent by the server;

and the second display module is used for displaying the retrieval result.

In another embodiment of the client provided in this specification, the client further includes:

the information input interface display module is used for displaying an information input interface, and the information input interface comprises an information input area, a first selection list and/or a second selection list; the information input area is used for inputting information; the first selection list and the second selection list are used for information selection, and the first selection list comprises category information of material application or material structures; the second selection list comprises category information of entity types;

and the input information acquisition module is used for acquiring input information based on the information input area, the first selection list and/or the second selection list.

On the other hand, an embodiment of the present specification further provides a method for generating summary information, which is applied to a server, and includes:

receiving a summary information generation request sent by a client, wherein the summary information generation request comprises a data source of a summary to be generated;

extracting entity information and incidence relation information between the entity information from the data source, and correlating the corresponding entity information according to the extracted incidence relation information; wherein, the incidence relation information comprises information describing the incidence direction and the incidence type between the entity information;

and generating summary information according to the associated entity information, and sending the generated summary information to the client so as to display the client.

the fifth receiving module is used for receiving a summary information generation request sent by the client, wherein the summary information generation request comprises a data source of a summary to be generated;

the eighth extraction module is used for extracting the entity information and the incidence relation information between the entity information from the data source; wherein, the incidence relation information comprises information describing the incidence direction and the incidence type between the entity information;

the sixth association module is used for associating the corresponding entity information according to the extracted association relation information;

the generating module is used for generating summary information according to the associated entity information;

and the seventh sending module is used for sending the generated summary information to the client so as to display the client.

On the other hand, an embodiment of the present specification further provides a summary information generating method, which is applied to a client, and includes:

sending a summary information generation request to a server, wherein the summary information generation request comprises a data source of a summary to be generated; so that the server extracts the entity information and the incidence relation information between the entity information from the data source, and correlates the corresponding entity information according to the extracted incidence relation information; wherein, the incidence relation information comprises information describing the incidence direction and the incidence type between the entity information; generating abstract information according to the associated entity information, and sending the generated abstract information to the client;

and receiving and displaying the summary information sent by the server.

the eighth sending module is used for sending a summary information generation request to the server, wherein the summary information generation request comprises a data source of a summary to be generated; so that the server extracts the entity information and the incidence relation information between the entity information from the data source, and correlates the corresponding entity information according to the extracted incidence relation information; wherein, the incidence relation information comprises information describing the incidence direction and the incidence type between the entity information; generating abstract information according to the associated entity information, and sending the generated abstract information to the client;

the fifth receiving module is used for receiving the summary information sent by the server;

and the third display module is used for displaying the summary information.

According to the information flow extraction method, the device and the equipment provided by one or more embodiments of the specification, the entity information and the incidence relation information between the entity information are extracted simultaneously, the extracted entity information can be further verified and screened by effectively utilizing the incidence relation information, whether the extracted entity information is the entity information related to the target entity information is determined, and therefore the accuracy of extracting the entity information related to the target entity information is further improved, and noise is effectively filtered. Meanwhile, the extracted entity information can be associated with the target entity information by utilizing the association relationship information, the association relationship between the extracted entity information and the target entity information is effectively displayed, the user can conveniently check and comb the association relationship, the user can accurately and efficiently obtain the required useful information, the required or new solution is found, and the use experience of the user is improved.

Drawings

In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort. In the drawings:

fig. 1 is a schematic flow chart of an information flow display method provided in the present specification;

FIG. 2 is a schematic illustration of information flow in one embodiment provided herein;

FIG. 3 is a schematic illustration of information flow in another embodiment provided herein;

FIG. 4 is a schematic representation of an information map in another embodiment provided herein;

FIG. 5 is a schematic representation of an information map in another embodiment provided herein;

FIG. 6 is a schematic representation of an information map in another embodiment provided herein;

FIG. 7 is a schematic representation of an information map in another embodiment provided herein;

FIG. 8 is a schematic representation of an information map in another embodiment provided herein;

FIG. 9 is a schematic representation of an information map in another embodiment provided herein;

FIG. 10 is a schematic representation of an information map in another embodiment provided herein;

FIG. 11 is a schematic representation of an information map in another embodiment provided herein;

FIG. 12 is a schematic representation of an information map in another embodiment provided herein;

FIG. 13 is a schematic representation of an information map in another embodiment provided herein;

FIG. 14 is a schematic illustration of an information atlas visualization in another embodiment provided by the present description;

FIG. 15 is a schematic illustration of a material property classification in another embodiment provided herein;

FIG. 16 is a schematic diagram of linking material to a list of attributes from different attribute classes using units and metrics in another embodiment provided by the present specification;

FIG. 17 is a schematic diagram of the extraction of attributes, units, and measures in another embodiment provided by the present specification;

FIG. 18 is a schematic illustration of linking extracted attributes, units, and measures into corresponding materials in another embodiment provided herein;

FIG. 19 is a schematic diagram of the extraction of attributes, units, and measures in another embodiment provided by the present specification;

FIG. 20 is a schematic diagram of the extraction of attributes, units, and measures in another embodiment provided by the present specification;

FIG. 21 is a schematic diagram of the extraction of attributes, units, and measures in another embodiment provided by the present specification;

FIG. 22 is a schematic illustration of linking extracted attributes, units, and measures into corresponding materials in another embodiment provided herein;

FIG. 23 is a flow chart illustrating a method for extracting information streams provided herein;

fig. 24 is a schematic diagram of a model structure of an information flow extraction apparatus provided in the present specification;

FIG. 25 is a flow chart illustrating an information retrieval method provided herein;

FIG. 26 is a schematic illustration of a search interface and search results in another embodiment provided by the present specification;

FIG. 27 is a schematic view of a search interface in another embodiment provided by the present specification;

FIG. 28 is a schematic diagram of the results of a search in another embodiment provided herein;

FIG. 29 is a schematic view of application classification in another embodiment provided by the present specification;

FIG. 30 is a schematic illustration of a material structure classification interaction interface in another embodiment provided by the present specification;

FIG. 31 is a schematic diagram illustrating the classification of polymers into further refinement of material structure in another embodiment provided herein;

FIG. 32 is a schematic illustration of an application classification interaction interface in another embodiment provided by the present specification;

FIG. 33 is a schematic diagram of a further refinement of the classification of smart materials in another embodiment provided by the present specification;

FIG. 34 is a schematic diagram of the results of a search in another embodiment provided herein;

FIG. 35 is a schematic view of a search interface in another embodiment provided by the present specification;

FIG. 36 is a schematic diagram of the results of a search in another embodiment provided herein;

FIG. 37 is a graphical representation of the protein structure, DNA plasmid, and material microstructure in another embodiment provided by the present specification;

FIG. 38 is a pictorial representation of a circuit diagram, flow diagram in another embodiment provided by the present description;

FIG. 39 is a schematic illustration of a material property classification in another embodiment provided herein;

fig. 40 is a flowchart illustrating a digest generation method provided in this specification.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in one or more embodiments of the present specification will be clearly and completely described below with reference to the drawings in one or more embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the specification, and not all embodiments. All other embodiments obtained by a person skilled in the art based on one or more embodiments of the present specification without making any creative effort shall fall within the protection scope of the embodiments of the present specification.

In one scenario example, a user may conduct a search for information through a client. The client may be a removable device. For example, the client may be a smart phone, a tablet electronic device, a portable computer, a Personal Digital Assistant (PDA), an in-vehicle device, or a smart wearable device. The client can also be a desktop device. For example, the client may be a server, an industrial personal computer (industrial control computer), a Personal Computer (PC), a kiosk, a smart self-service terminal (kiosk), or the like.

The user may enter information in a page presented by the client. The input information is reference information which a user wants to search for information. If the user wants to inquire the information related to the alumina ceramics, the user can input the information alumina ceramics in the page displayed by the client, and the client can send an information search request to the server, wherein the search request can be accompanied by the input information alumina ceramics. The server can extract input information alumina ceramic from the information search request, and extract entity information or data sources which are associated with the input information alumina ceramic by taking the input information alumina ceramic as reference information.

The server may have a memory or information linked to a database. After receiving the information search request sent by the client, the server may perform information search by using the data source in the server memory or the data source in the database linked to the server memory. After the server finishes information search, the searched information can be fed back to the client side, so that the client side can display the searched information.

However, generally, information may have a plurality of different description forms in the same data source or different data sources, and even a large amount of incomplete and non-descriptive information exists, so that the searched information has the problems of inaccuracy and incompleteness. The searched information amount may be huge, irrelevant information or more noise information is provided, and the information cannot be simply displayed to the user, so that the user needs to expend more energy to screen and comb when checking, the user easily misses the desired information, and the experience is poor.

In some embodiments, the information stream corresponding to the input information may be formed by extracting associated entity information of the input information and associated relationship information such as an association type and an association direction between the input information and the associated entity information from a data source, and associating the extracted associated entity information with the input information by using the associated relationship information. Then, the information flow can be displayed to the user, so that the needed information and the association relation between the information can be visually checked in the information flow mode. In a specific embodiment, as shown in fig. 1, in an embodiment of an information flow presentation method provided in this specification, the method may include:

s20: the client sends an information flow acquisition request to the server, wherein the information flow acquisition request comprises input information acquired by the client.

The client may send an information stream acquisition request to the server. The information flow obtaining request may include the input information of the user obtained by the client. For example, the client may correspond to an information input box, and a user may input information into the client by manually inputting or selecting an information pull-down selection box displayed by the client. Then, the client may further respond to the operation of the user to generate an information flow obtaining request, where the information flow obtaining request may be accompanied with the input information. The client may then send an information flow acquisition request to the server. For example, a search button or an information stream generation button may be arranged in a display interface of the client, a user may trigger a search instruction by clicking the search button or by clicking an enter key in a keyboard, and then the client may generate an information stream acquisition request in response to an operation of the user and send the information stream acquisition request to the server. Of course, the specific implementation manners of the information input, the user operation, and the information flow obtaining request are only examples, and the specific implementation is not limited to the above manners.

In one example scenario, the client may present the user with an information input box in which the user may input the information "polyvinyl alcohol" and then click on the information flow generation button, and may generate an information flow acquisition request to acquire all or part of the information about the material in response to the click operation of the user. Alternatively, the user inputs the information "organic electroluminescent display coating" in the input box to acquire all or part of the information from the raw material to the material application with respect to the material application. Alternatively, other types of information may be input to obtain information related to the input information.

In another example scenario, the user may also define the extraction direction of the desired information flow or the start node and the end node of the information flow. Such as to define a process information stream for extracting a material to obtain material process related information for the material. The user can input the 'polyvinyl alcohol material process' or the 'polyvinyl alcohol and processing process' and the like in a page displayed by the client. Or, a page displayed by the client is also provided with a pull-down selection box, and one or more information extraction direction selection items, such as material technology, material application and other options, are arranged in the pull-down selection box. After the user inputs "polyvinyl alcohol", the user can further select an information extraction direction from the pull-down selection box to obtain corresponding information. Accordingly, "polyvinyl alcohol" and the extraction direction "processing" may be accompanied by a search request and sent to the server.

S22: and the server receives the information flow acquisition request and extracts one or more entity information from the input information as target entity information.

After receiving the information flow acquisition request, the server may extract one or more pieces of entity information from the input information in the information flow acquisition request, and extract the information flow as target entity information.

The entity information may include information such as a proper name or meaningful number of phrases identified from the data source. Such as may include material names, specific material applications, manufacturing methods, material properties, trade names, and the like. A quantitative phrase of the meaning such as a measure of a material property, a date, etc. The entity information may include an entity type and an entity value. The entity type may refer to information that classifies information having common characteristics and describes a category to which the information belongs. The entity value may include specific information corresponding to each entity type in the data source. The data sources may include text information such as relatively closed patent documents, paper documents, etc., as well as text information in open source databases, etc. The information in the data source is huge and the information type is complex, and the workload of information analysis and extraction is usually large. By combining the entity information to extract the information, the accuracy and the efficiency of information extraction can be greatly improved.

For the material technology area, the entity types may include, for example, a material name entity, a material structure type entity, a material application entity, a material process entity, a material property entity, a material trade name entity, a material supplier/manufacturer entity, a unit entity, a measurement entity, and the like. The material process entity may further include a process method entity between materials, a manufacturing method entity between materials and material applications, an intermediate material entity in a processing or manufacturing process, and the like. The intermediate material entities may for example comprise additive entities, catalyst entities etc.

The material name entity may include an entity type corresponding to the description information of the material name to determine what kind of material. The entity value corresponding to the material name entity can be chemical, physical or biological and the like with name description information of materials, such as polyvinyl alcohol, zirconium oxide, martensitic stainless steel and the like.

The material structure type entity may include an entity type corresponding to description information of a type to which each material belongs, which is determined according to a material structure. The material structure type entity can be metal, ceramic, biological and the like. The material application entity may include an entity type corresponding to description information of an application to which the material corresponds. The entity value of the material application entity can be the application field or specific application information of each material. The application fields may include, for example, building materials, energy materials, 3D printing materials, optoelectronic materials, and the like. Specific applications may include, for example, organic electronic display coatings, packaging bags, and the like.

The material process entity may include an entity type corresponding to one or more materials from which description information for another one or more materials or processes/manufacturing methods to which the materials are applied is obtained. The entity value of the material process entity may be, for example, descriptive information of the process/manufacturing method, such as polymerization, hydrolysis, alloying, metal refining, drawing, finishing, etc., and additives, catalysts, etc. in the material processing or manufacturing process.

The material attribute entity may include an entity type corresponding to description information characterizing intrinsic characteristics of the material, such as structure, performance, and the like. The physical values of the material process entity may include, for example, melting point, thermal conductivity, specific heat capacity, yield strength, elastic modulus, and the like.

The material trade name entity can comprise an entity type corresponding to trade name description information used in the actual production and sale process of the material. The material supplier/manufacturer entity may include an entity type corresponding to the supplier/manufacturer description information of each material.

The unit entity may include an entity type corresponding to the measurement unit description information of the specific value of the attribute information of each material. The measurement entity may include an entity type corresponding to the specific value description information of the attribute information of each material. If the unit of the melting point may include celsius degree, kelvin, etc., the corresponding celsius degree and kelvin are the entity values corresponding to the unit entity. For example, tungsten has a melting point of 3410 degrees celsius, where melting point is the value of the entity that is a property of the material, 3410 is the value of the entity that is measured, and degrees celsius is the value of the entity in units of the entity.

For the input information "polyvinyl alcohol", the entity information that can be extracted is "material name entity-polyvinyl alcohol". In A-B in the entity information expression form, A is an entity type, B is an entity value, namely the material name entity is the entity type, and polyvinyl alcohol is the entity value. The expression manner of the entity type and the corresponding entity value is only an example, and a correlation or a table storage manner may be adopted during specific data processing. For convenience of unified description, entity information in the embodiments of the present specification is uniformly in the form of "a-B". For example, for the above "melting point of tungsten is 3410 degrees celsius", the entity information that can be extracted is "material name entity-tungsten", "material property entity-melting point", "unit entity-degree celsius", "measurement entity-3410".

In some embodiments, the method for extracting entity information may be obtained by performing learning training on a large amount of sample data, for example, a method for extracting named entities may be used. In some embodiments, the input information may be first processed and tagged using Natural Language Processing (NLP) algorithms, for example. Then, Entity information in the input information may be extracted using a Named Entity recognition algorithm (NER) based on a pre-constructed Entity information dictionary or Entity information base. The entity information dictionary may include common professional word information in a specific field pre-constructed based on a large amount of sample data, and may be, for example, a database including professional word information such as material names, material attributes, material applications, and the like in a material field. The entity information base may include a database pre-constructed by the platform and including general professional word information of a specific field, such as an entity information base including professional word information of material names, material attributes, material applications, and the like of a material field. By extracting the entity information based on the pre-constructed entity information dictionary or the entity information base, the efficiency and the accuracy of extracting the entity information can be greatly improved. The server may extract the information flow using the extracted entity information as target entity information. The target entity information is reference information extracted from the information stream, and the server may extract entity information associated with the target entity information and association relationship information between the target entity information and the extracted entity information, based on the target entity information, to construct an information stream for obtaining corresponding target entity information. For example, for the input information mentioned in the above example, the entity information mentioned therein is "material name entity-polyvinyl alcohol", and the entity information can be used as target entity information to extract entity information associated with "material name entity-polyvinyl alcohol" and association relationship information between "material name entity-polyvinyl alcohol" and the extracted entity information from a data source, thereby constructing an information stream for obtaining "material name entity-polyvinyl alcohol".

In some embodiments, if more than one entity information is extracted from the input information. The server may further analyze the information flow extraction requirements corresponding to each entity information extracted from the input information. If the extraction requirement is to extract the information streams corresponding to the entity information respectively, the server may take the entity information as the target entity information respectively to extract the information streams corresponding to the entity information respectively. For example, the input information is "polyvinyl alcohol OR polycarbonate", the input information is in the form of boolean logic expression, and if the meaning of the logical operator "OR" is predefined, the server may determine the information flow extraction requirements of the two entity information to be respectively extracted according to the related word "OR". The server may use "material name entity-polyvinyl alcohol" and "material name entity-polycarbonate" as the target entity information, respectively, that is, there are two target entity information, and the contents of the two target entity information are "material name entity-polyvinyl alcohol" and "material name entity-polycarbonate", respectively.

Then, the server can extract the information streams of the two target entity information "material name entity-polyvinyl alcohol" and "material name entity-polycarbonate" respectively to obtain the information stream of the "material name entity-polyvinyl alcohol" and the information stream of the "material name entity-polycarbonate". In specific implementation, the server may sequentially and respectively extract information streams of each target entity information, or may extract information streams of a plurality of target entity information in parallel by means of parallel processing.

If the extraction requirement is direction-limited extraction, the extraction direction limitation is included in the input information as mentioned in step S20. The user inputs "polyvinyl alcohol" in the input box and selects or inputs the extraction direction definition information "organic electroluminescent display coating" in the extraction direction definition selection box or the input box. The server extracts information streams composed of information of material processing processes between the material polyvinyl alcohol and the material application organic electroluminescent display coating, other materials involved, material suppliers, and the like based on the direction definition extraction requirement after extracting two entity information of 'material name entity-polyvinyl alcohol' and 'material application entity-organic electroluminescent display coating' from the input information.

Of course, the above information flow extraction request, the extraction request selection method, the input method, the interface display method, and the like are preferred examples, and when the information flow extraction request is specifically implemented, other methods may be used for implementation, and are not limited herein.

S24: the server extracts at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information.

For any one piece of target entity information, the server can extract at least one piece of associated entity information corresponding to the target entity information from the data source and extract the association relation information between the target entity information and the associated entity information. The associated entity information may refer to entity information having a certain association relationship with the target entity information. The association relation information may include information describing an association direction and an association type between the entity information.

The association direction can represent the circulation direction of the entity information in the information flow so as to visually identify the flow direction of the information flow, and the entity information is convenient to comb and view. If the material monomer is subjected to a material process to obtain the material polymer, the correlation direction among the material monomer, the material process and the material polymer is that the material monomer is subjected to the material process to obtain the material polymer. In some embodiments, when the information flow is presented, the association direction may also be represented by an arrow, as shown in fig. 2, where the arrow in fig. 2 identifies the association direction between the entity information in the information flow.

The association type can comprise a characteristic description of the association relationship between entity information with different association relationship characteristics, so as to better comb and verify whether and what kind of association exists between the entity information. Such as material-to-material via a machining process, the corresponding type of association may be material-to-material. The type of association between the material and the material type information is material and material type.

The association direction and the association type between the entity information can be determined by analyzing the text information where the entity information is located. For example, a large amount of sample data can be trained, and the association direction and the association type between entity information can be extracted from text information by using an algorithm model obtained by training. In some embodiments, the association direction and the association type between the entity information may also be determined by comprehensively considering the entity type between the entity information, so as to determine the association direction and the association type more accurately and conveniently.

The association relation information may include information describing an association direction and an association type. The information may be described in a plain text, for example, by associating the type of material with the material type and associating the direction of the material type with the synthetic resin being polyvinyl alcohol. Alternatively, an information linking mode may be adopted, for example, the server establishes a link for two or two entity information through a symbol, a graph, and the like, and then marks the association type and the association direction on the link. Of course, the association type and the association direction may be represented by numbers, symbols, figures, and the like. Of course, other information description manners may be adopted.

The association relationship information may include information describing the association direction and the association type, or may only include information describing the association direction or the association type. For example, for entity information between name variants of material names, there may be no information describing the associated direction, and there is only an associated type to indicate that the two are co-located information. For entity information of the material corresponding to the material type, only description information of the associated type can be related. For the entity information corresponding to the process between the materials, only the description information of the related direction may be used. For the entity information only having the association type, the association type may be marked when the information stream is displayed, or the entity value and the entity type may also be marked when the entity information is displayed, the association relationship between the entity information is represented by the entity type, and meanwhile, the entity information may be associated through a bracket or a solid line without an arrow. And for the incidence relation information comprising the incidence direction, the information circulation direction between the entity information can be represented by using a straight line with an arrow according to the incidence direction in the incidence relation information so as to be conveniently and visually displayed. Of course, the display mode of the association direction and the association relationship is only an example, and other modes may be adopted in specific implementation.

In some embodiments, the associated entity information and the association relationship information corresponding to the target entity information may be determined by analyzing context information of the target entity information in the data source.

The context information may include a statement in which the target entity information is located and one or more statements before and after the statement in which the target entity information is located. Alternatively, if there is a strong association relationship between the paragraph in which the target entity information is located and the paragraphs before and after the paragraph, the context information may also include the paragraph in which the target entity information is located and one or more paragraphs before and after the paragraph.

In some embodiments, for example, a statement in which the target entity information is located may be initially taken as context information in which the target entity information is located to perform extraction of the associated entity information and the association relationship information between the associated entity information and the associated entity information. Then, one or more sentences before and after the sentence where the sentence is located can be further supplemented into the context information, and the extracted associated entity information and the association relation information between the associated entity information and the associated entity information are corrected by using one or more sentences before and after the sentence where the sentence is located. Alternatively, one or more paragraphs before and after the paragraph may be further supplemented to the context information, and the extracted associated entity information and the association relationship information between the extracted associated entity information and the associated entity information may be corrected by using one or more paragraphs before and after the paragraph.

Or, the statement in which the target entity information is located may be initially taken as context information in which the target entity information is located to extract the associated entity information and the associated relationship information between the associated entity information, when the associated entity information and the associated relationship information between the associated entity information cannot be extracted, one or more statements before and after the statement in which the target entity information is located may be further supplemented to the context information, and the associated entity information of the target entity information and the associated relationship information between the associated entity information may be further extracted by using one or more statements before and after the statement in which the target entity information is located. Of course, one or more paragraphs before and after the paragraph may be further supplemented to the context information, and the associated entity information of the target entity information and the association relationship information between the associated entity information may be further extracted or corrected by using one or more paragraphs before and after the paragraph.

Generally, entity information with strong association relation can be described in a sentence, and the sentence where the target entity information is located is primarily used as context information for information extraction, so that interference of other information can be avoided, and associated entity information corresponding to the target entity information and association relation information between the associated entity information and the associated entity information can be extracted more accurately and efficiently. And then, the incidence relation information between the extracted entity information is further corrected by further combining the preceding and following sentences or the preceding and following paragraphs, or the further extraction is realized when the entity information or the incidence relation information between the entity information is not extracted, so that the accuracy and the comprehensiveness of the extraction of the entity information and the incidence relation between the entity information can be greatly improved.

In some embodiments, the data source where the target entity information is located may be retrieved first, and then the context information of the target entity information in the data source may be located. Then, the context information where the target entity information is located may be analyzed by using a natural language processing algorithm, and the like, and associated entity information corresponding to the target entity information and association relationship information between the associated entity information and the target entity information are extracted.

The data source where the target entity information is located can be initially retrieved through methods such as information matching. The context information of the target entity information in the data source may then be located using the methods provided in the embodiments described above. The entity information can be extracted from the context information of the target entity information by using a named entity extraction method, and then the sentences of the extracted entity information and the target entity information or the sentences of the extracted entity information and the sentences of the target entity information are analyzed by using semantic role labeling, dependency semantic analysis, part of speech labeling and other methods, so as to extract the description information which can describe the association direction and the association type between the extracted entity information and the target entity information. The extracted description information may be analyzed, or the extracted description information, the extracted entity information, and the target entity information may be comprehensively analyzed, and if a certain association direction and/or association type exists between the extracted entity information and the target entity information, the extracted description information may be used as the associated entity information, and the extracted description information may be used as the association relationship information.

For example, for a certain statement that the target entity information "polyvinyl alcohol" is in the data source, "polyvinyl alcohol is a synthetic resin, and is usually prepared by alcoholysis (usually called hydrolysis or saponification) of polyvinyl acetate," the entity information "material name entity-polyvinyl alcohol," material type entity-synthetic resin, "" material name entity-polyvinyl acetate, "" material process entity-alcoholysis, "" material process entity-hydrolysis, "and material process entity-saponification" can be extracted by using the named entity extraction algorithm.

The method can further perform semantic role labeling, dependency semantic analysis, part of speech labeling and other natural language processing methods on the entity information and the words between the entity information in the sentence, extract the association relation description information ' X ' between the ' polyvinyl alcohol ' and the ' synthetic resin ' as a kind of Y ', further combine the entity types of the ' polyvinyl alcohol ' and the ' synthetic resin ' to determine that the association type between the two is ' material and material type ', and the association direction is ' synthetic resin ' as the material type of the ' polyvinyl alcohol '.

The association of "polyvinyl alcohol", "polyvinyl acetate" and "alcoholysis" describes the information "X was prepared from Y for Z", by further analyzing the entity types of the three entities: the alcoholysis corresponds to a material process entity, the entity type corresponding to the polyvinyl alcohol and the polyvinyl acetate is a material name entity, the association type between the three entities can be further determined to be a material-to-material process, and the association direction is that the polyvinyl acetate obtains the polyvinyl alcohol through alcoholysis.

Meanwhile, the description information of the association relationship between "alcoholysis", "hydrolysis" and "saponification" (usually called hydrolysis or saponification) "can be extracted, and through the description information and the entity types of" alcoholysis "," hydrolysis "and" saponification ", the association type between the three can be determined to be" co-located entity ", the association direction is co-located information, and the co-located entity or co-located information represents other expression forms that each entity information in two or more entity information is other entity information.

The processing natural language processing methods such as a named entity extraction algorithm, semantic role labeling, dependency semantic analysis, part of speech labeling and the like used in the extraction process of the entity information and the incidence relation information can be obtained by deep learning a large amount of sample data according to the extraction characteristics of the entity information and the incidence relation information corresponding to the information flow extraction, so that the extraction accuracy of the entity information and the incidence relation information can be improved.

In other embodiments, the associated entity information and the association relationship information corresponding to the target entity information may be determined in the following manner:

for the context information of the determined target entity information, the context information may be processed by using an NLP algorithm, for example, the data source may be processed and labeled by using the NLP algorithm. The NER may then be utilized to identify entity information in the context information based on an entity information dictionary or an entity information base. For example, for the construction of the material information stream, the entity information included in the context information may be extracted from the context information by using an NER algorithm based on an entity information dictionary or an entity information library corresponding to the material, or an entity information dictionary or an entity information library corresponding to the material structure type corresponding to the target entity information.

For example, if the target entity information is "polyvinyl alcohol", and the material structure type corresponding to the "polyvinyl alcohol" is a polymer, an entity information dictionary or an entity information library corresponding to the polymer is obtained. And extracting the entity information from the context information of the target entity information by using an NER algorithm, and comparing the extracted entity information with information in an entity information dictionary or an entity information base, so that the entity information in the context information can be identified more accurately and efficiently. Such as may identify corresponding descriptive information for the monomer, polymerization reaction, polymerization medium, polymer class, polymer type, manufacturing process, manufacturer, supplier, chemical modification, physical modification, application, and the like.

Then, the NLP algorithm filtering method can be used for removing the interference information in the extracted entity information. The interference information may include stop words, general words, non-scientific words, words unrelated to material context, and the like. Then, taking the extracted entity information and the target entity information as objects, and further analyzing the context information by using an NLP algorithm for each object to determine the metadata corresponding to each object. The metadata may refer to information describing each object. Then, the metadata corresponding to the target entity information and the metadata corresponding to the extracted entity information may be compared with each other, and whether an association relationship and specific association relationship information exist between the target entity information and the extracted entity information may be determined. If the target entity information and the extracted entity information have an association relationship, the entity information having the association relationship may be used as the association entity information of the target entity information, and the association relationship information between the two may be determined based on the comparison result to associate the two.

S26: and the server associates the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information.

The server can associate the corresponding target entity information and the associated entity information by using the extracted association type and association direction, and can obtain an information stream corresponding to the target entity information. The related implementation may be performed with reference to the content described in step S24, and is not described herein again.

S28: and the server sends the information flow to the client side so as to display on the client side.

The server may send the information stream to the client for presentation on the client. If the information flow can be visualized by using a visualization tool, the visualized information flow is sent to the client to be visually displayed on the client. Fig. 2 shows the information flow established by the above example.

By simultaneously extracting the entity information and the incidence relation information between the entity information, the extracted entity information can be further verified and screened by effectively utilizing the incidence relation information to determine whether the extracted entity information is the entity information related to the target entity information, so that the accuracy of extracting the entity information related to the target entity information is further improved, and the noise is effectively filtered. Meanwhile, the extracted entity information can be associated with the target entity information by utilizing the association relationship information, the association relationship between the extracted entity information and the target entity information is effectively displayed, the user can conveniently check and comb the association relationship, the user can accurately and efficiently obtain the required useful information, the required or new solution is found, and the use experience of the user is improved.

For example, a user is trying to find information about part or all of a processing technology or a manufacturing method, a raw material, an additive, etc. of a specific material or a material application, and associated entity information about the specific material or the material application in the same or different data sources is extracted and associated through the scheme provided by the above embodiment to form an information stream, which is displayed to the user. By utilizing the information flow corresponding to the specific material, a user can simply and conveniently find the required information, the process that the user performs line screening and combing on a plurality of different data sources is avoided, and the time is saved. Meanwhile, through the information flow mode, information of each entity can be displayed to a user in an associated mode, so that the omission of information and the omission of the information caused by large data volume or unknown industry during information screening by the user can be avoided, and the association between the information can be avoided, so that the user can more accurately obtain the needed solution and the information such as raw materials and processing technologies needed in the solution. In addition, by the method for simultaneously extracting the entity information and the incidence relation information between the entity information, the extraction of the noise information in the data source to the effective information can be effectively eliminated, and the accuracy of the information obtained by the user is improved.

The information in the data source is complex and variable, and the entity information and the incidence relation information between the entity information are generally difficult to accurately extract. In some embodiments, an algorithm model of associated entity information corresponding to the target entity information and associated relationship information between the associated entity information may also be trained based on the entity type of the target entity information, so as to further improve the accuracy of extracting the associated entity information and the associated relationship information between the entity information. For the entity information in the material technical field, the material-to-material or material-to-material application is usually converted to-material or material-to-material application by a certain processing technology or manufacturing method, and a material or material application is obtained by which material is processed, and by which processing technology or manufacturing method, it is usually also the user's desire to know. Based on the application scenario, in other embodiments of the present specification, the extracting at least one associated entity information corresponding to the target entity information and association relationship information between the target entity information and the associated entity information from a data source may include:

retrieving a data source where the target entity information is located;

locating context information of the target entity information in a data source;

The material process information correlation mode can be determined according to information correlation characteristics between materials or between materials and material applications. When the context information of the target entity information includes a material process entity, a material name entity and/or a material application entity, there may be a possibility that the context information generally relates to what kind of conversion of the processing technology one or more materials are generally obtained or what kind of manufacturing method the material application is obtained. When the material name entity, the material application entity, and the material process entity are extracted, the corresponding information association manner may generally be a manner in which the material name entity obtains the material name entity through the material process entity, or a manner in which the material name entity obtains the material application entity through the material process entity. In this case, the material process entity may be the material name entity or the associated entity information of the material application entity, or may be a part of the associated relationship information between a plurality of material name entities or between a material name entity and a material application entity.

In some embodiments, a large number of data sources related to the material process entity, the material name entity and/or the material application entity may be learned by using an algorithm such as deep learning, so as to accurately associate the material name entity or the material name entity and the material application entity with the material process entity, thereby accurately extracting entity information and an association relationship between the entity information.

For example, the user enters the information as polyvinyl alcohol in the input box of the client and clicks the information stream extraction button. The client can generate an information flow acquisition request based on the polyvinyl alcohol and send the information flow acquisition request to the server. After receiving the information flow acquisition request, the server may extract input information from the information flow acquisition request, and may further extract the input information. The server extracts entity information from the input information, and the extracted entity information may include "material name entity-polyvinyl alcohol". The server may then perform information extraction on the data sources in the database to extract one or more data sources comprising polyvinyl alcohol. The contextual information in which the polyvinyl alcohol is located can then be addressed.

Assume that the located context information is as follows: the "1, 2-ethanediol polyvinyl alcohols" can be produced according to known or customary production methods. For example, 1, 2-ethanediol polyvinyl alcohol the following production methods can be used: vinyl acetate polymerization under a pressure (load) higher than that under conventional conditions to produce polyvinyl alcohol; and copolymerization of vinyl acetate with ethylene carbonate to obtain 1, 2-ethanediol polyvinyl alcohol "in the above-mentioned content.

By using a named entity extraction method, a material name entity-vinyl acetate and a material process entity-polymerization can be extracted; and entity information such as "material name entity-vinyl acetate", "material name entity-vinyl carbonate", "material process entity-copolymerization", and the like. The entity information corresponding to the material name entity can be primarily used as the associated entity information, and the entity information corresponding to the material process entity can be used as the associated type between the material name entities. Meanwhile, the sentence where the information is located is analyzed to further determine the association direction. Analysis of the descriptors "polymerization of vinyl acetate to produce polyvinyl alcohol" and "copolymerization of vinyl acetate with vinyl carbonate to obtain 1, 2-ethanediol polyvinyl alcohol in the above-mentioned amounts" gives "polyvinyl alcohol" by "polymerizing" the "vinyl acetate" and "polyvinyl alcohol" by "copolymerizing" the "vinyl acetate" with the "vinyl carbonate" in the direction of attachment.

Then, the "polyvinyl alcohol" and the "vinyl acetate" may be associated according to the extracted association relationship information such as the association type, the association direction, and the like, and the "polyvinyl alcohol" and the "vinyl acetate, vinyl carbonate" may be associated according to the extracted association relationship information "copolymerization". If the association direction can be represented by an "arrow", the information flow formed after association is presented by an information presentation algorithm, as shown in (a) and (b) of fig. 3, respectively.

And performing information association on the extracted information comprising the material name entity, the material application entity and the material process entity based on the information association characteristics of the material process, so that the accuracy of extracting the association relation information between the material entity information and the material entity information can be further improved.

In other embodiments of the present description, the association relationship information may further include peer description information, and the peer description information may include description information describing that the associated entity information is in other expression forms of the target entity information. When the association relationship information includes peer description information, the server may supplement the associated entity information corresponding to the peer description information to the target entity information, and update the target entity information. Correspondingly, the server may extract at least one associated entity information corresponding to the updated target entity information and the association relationship information between the target entity information and the associated entity information from the data source. And associating the updated target entity information with the associated entity information by using the association relation information to obtain an information stream corresponding to the updated target entity information.

For example, with respect to the polyvinyl alcohol mentioned in the above example, when the data source is analyzed, the description information of "polyvinyl alcohol may also be expressed as" poly (vinyl alcohol) "or" PVA "is extracted, and the description information may be used as the association relation information to associate the poly (vinyl alcohol), the polyvinyl alcohol, and the PVA. Poly (vinyl alcohol) and PVA may also be supplemented to the target entity information and the target entity information may be updated, and accordingly, the updated target entity information may include polyvinyl alcohol, poly (vinyl alcohol), and PVA. Then, the data sources can be searched by using poly (vinyl alcohol), polyvinyl alcohol and PVA, and the related entity information and the related relationship information can be extracted by using poly (vinyl alcohol), polyvinyl alcohol and PVA in the searched data sources. After obtaining the sub information streams corresponding to the poly (vinyl alcohol), the polyvinyl alcohol, and the PVA, the information streams corresponding to the updated target entity information may be obtained by associating the sub information streams extracted based on the poly (vinyl alcohol), the polyvinyl alcohol, and the PVA with the poly (vinyl alcohol), the polyvinyl alcohol, and the PVA as reference information.

Of course, for other entity types, the above manner can also be adopted to associate different expression forms of the same entity information, and then further extract the entity information, such as "alcoholysis", "hydrolysis", and "saponification" in the above example belong to the collocated information corresponding to the extracted material process entity.

Or, in another embodiment, based on the scheme provided in the foregoing embodiment, metadata of each entity information may be extracted, the metadata corresponding to each entity information are compared with each other, and if it is determined that the metadata corresponding to the two entity information have a certain similarity or an explanation that the two entity information are opposite to each other is determined through metadata comparison, the two entity information may be a co-located entity. There may be many different descriptions of the same material, material application, process, manufacturing method, etc. in the data source, for example, polyvinyl alcohol may also be described as english spellings or other language expressions. The detailed and standardized descriptions of material names, applications or processes and the like are usually lacked in various complicated and huge data sources, and when a user searches information, the information input by the user is usually only in one or more forms known by the user, so that the finally inquired information is not accurate and comprehensive. For example, due to the different way in which material names are expressed in patents, a user searching for certain materials such as polyvinyl alcohol may not find certain patents that use PVOH, PVA, poly (vinyl alcohol), hydrolyzed polyvinyl acetate, etc., resulting in a user infringement analysis that is less than comprehensive and thus risking infringement.

By using the scheme of the embodiment, when other expression forms of the entity information are extracted during the extraction of the entity information, the other expression forms are updated to the reference information depended during the information search in time, and the updated reference information is used for searching and associating the information, so that the accuracy and comprehensiveness of the information extraction can be greatly improved. During specific display, a universal name can be provided or selected for multiple material name variants and is linked with each material name so as to identify the material name, and a user can conveniently comb and look up information.

In still other embodiments, the method may further comprise: the server extracts the target entity information and the trade name of the associated entity information from a data source; and extracting a manufacturer or a supplier corresponding to the target entity information or the associated entity information according to the trade name, and associating the manufacturer or the supplier with the target entity information or the associated entity information. By further extracting the trade names corresponding to the materials, extracting the supplier or manufacturer information corresponding to the materials from resources such as an open source database and the like through the trade names, and associating the supplier or manufacturer information with the materials, a user can conveniently and directly obtain the manufacturing source of the materials according to information flow, and the use experience of the user is improved.

Multiple associated entity information of the target entity information and association relation information between the target entity information and different associated entity information can be extracted from the same data source or multiple data sources. In some embodiments, the server may further use an information stream formed by associating one or more extracted associated entity information with corresponding association relationship information as a sub-information stream, and then link different sub-information streams extracted from the same data source or different data sources by using the target entity information as reference information to obtain an information stream corresponding to the target entity information, and then display the final information stream to the user, thereby further improving the association between the information and facilitating the user to perform comprehensive analysis and viewing.

Suppose that a user is looking for a through solution for a particular material, e.g., from monomer to application. Through the information flow, a user can stitch together entities obtained from different sources (databases, patents, and the like), so that in order to obtain a complete solution at a single position, analysis on a plurality of different data sources is not needed, and time is saved.

Alternatively, the user needs to search for a manufacturing process for a new material, but cannot search because the patent for the manufacturing process has not yet been published or filed. However, there may already be journal articles, news articles, or other types of data sources that describe the material process. By extracting the information in the multiple data sources and performing correlation fusion, the user can more accurately and intuitively acquire the existing manufacturing process.

For example, the information streams shown in (a) and (b) in fig. 3 may be regarded as sub information streams, and the information stream obtained by integrating two sub information streams is shown in (c) in fig. 3. Based on the foregoing solution, in some embodiments, the associating the target entity information and the associated entity information by using the association relationship information may include: acquiring at least one associated entity information extracted from a data source and associated relationship information between the target entity information and the corresponding associated entity information, and associating the target entity information with the corresponding associated entity information by using the associated relationship information between the target entity information and the corresponding associated entity information to obtain a sub information stream of the target entity information; and linking sub information streams obtained from the same or different data sources by taking the target entity information as reference information to obtain an information stream corresponding to the target entity information.

In other embodiments, the method further comprises: the server can also extract information streams by taking each associated entity information extracted from the data source as target entity information to obtain information streams corresponding to a plurality of target entity information; and linking information streams corresponding to the target entity information by using the corresponding target entity information as reference information to obtain an information map.

The server can further extract information streams by taking the extracted associated entity information as target entity information, and further extract information streams corresponding to the associated entity information, so that a series of information streams can be obtained. And linking each information flow by taking the information of each associated entity as reference information, so that an information map can be obtained. Fig. 4 is an information map formed by the whole workflow from vinyl acetate (monomer) to organic electroluminescent display coating (application) associated with polyvinyl alcohol obtained by a user searching for polyvinyl alcohol, wherein the content in each box in fig. 4 is entity information, and an arrow indicates the associated direction.

The information flow is further extracted by utilizing the associated entity information, and then the information map is constructed by further linking, so that all information from the material monomer to the final application can be obtained, and a user can conveniently and comprehensively, accurately and visually know the materials, the applications, the processes and the manufacture related in the whole life cycle of a certain material.

As with the above-described embodiment, a user can easily identify the entire process information flow from monomer to chemically modified polymer (including catalyst, polymerization, chemical modification, physical modification, and additives) in the data source by means of the information flow, thereby also helping the user to find alternative polymerization processes, additives, catalysts, and the like.

For example, a user who normally searches for polyethylene may not get a document that labels polyethylene with PE, polyethylene HDPE, or industry polyethylene trade names. He may get incomplete information on different aspects of the material details, e.g. some polyethylene patents may contain monomers but no information on the polymerization process, the polymer treatment performed on the polymer or even the different applications in which the polyethylene can be used. The user may need to search different patents to search for different details of the polyethylene. With the above scheme, a user can get all information of a specific material through a single search, including name variations, suppliers, material applications, manufacturing processes, and the like.

Meanwhile, by utilizing the information flow of the embodiment, a user can simply and conveniently find the appropriate material for a specific application. By using information flows, users can discover new materials, suppliers, and material manufacturing techniques for specific applications. For example, a user looking for a material suitable for use in a medical stent may find a material such as Bioflow-V, Orsio (trade name). However, the material behind the trade name is not available from the same source. By using the information map formed by interlinking the above information, the user can know that Orsio uses cobalt chromium alloy as the base metal and has knowledge of the active poly L-lactide (PLLA) polymer coating. Other materials for making stents may be further explored by the user, including stainless steel with poly (ethylene-vinyl acetate) or poly (n-butyl methacrylate) coatings (CYPHER stents), cobalt chrome with carbon coatings (CRE8 stents), and the like. The user may then also use the information map to obtain the entire manufacturing process for each material used in the stent.

Fig. 4 to 13 are schematic diagrams of information maps of several common material types given in the present specification. Wherein, fig. 4 and 5 are information map schematic diagrams of the polymer material. Fig. 6 and 7 are schematic diagrams of information maps of the metal material. Fig. 8 and 9 are schematic diagrams of information maps of the ceramic material. Fig. 10 and 11 are schematic diagrams of information maps of biomaterials. Fig. 12 and 13 are schematic diagrams of information maps of materials. Fig. 4, 6, 8, 10, and 12 are information maps formed by associating entity information extracted from a data source with entity information. Fig. 5, 7, 9, 11 and 13 are schematic diagrams of the above five materials in general information flow.

Comparing the five materials with the information graph obtained after the actual information is extracted from the data source and the general information flow schematic diagram, according to the scheme of the embodiment of the specification, a user can input the entity information corresponding to any one or more nodes in the information graph, such as a single material, material application, material process and the like, then obtain part or all information related to the input information and the flow relation among the information, and finally obtain the information flow through correlation. The user can accurately obtain partial or whole information flow of each material from the material monomer to the material application, and meanwhile, information such as name variation, processing technology, manufacturing method, manufacturer, supplier and the like is effectively related. The method and the device effectively realize that the user accurately and comprehensively obtains part or all information related to the input information of the user and the circulation relationship between the information through less input information.

Of course, the data sources of the information nodes in the information stream may be further associated, as shown in fig. 4, 6, 8, 10, and 12, where different gray levels represent entity information from different data sources. In practical applications, different data sources can be represented by different colors, or each of the information nodes can be clicked to check from which data source each of the information nodes comes. Through further association and display of the data sources, information from different data sources can be effectively distinguished at a single node, so that the method is beneficial for a user to effectively control technologies owned or utilized by different patentees, and the analysis of infringement, patent rights and the like is carried out more simply and conveniently. Based on the application scenario, in other embodiments, the sub-information stream of the target entity information may be further associated with a corresponding data source. By associating the extracted sub-information streams with the corresponding data sources, a user can conveniently check the sources of the corresponding sub-information streams, and the tracking of each piece of information in the information streams is effectively realized.

In other embodiments, based on the solutions provided in the above embodiments, the page displayed by the client may further include options of a start node and an end node of the information stream, and the user may click the corresponding options and correspondingly input the corresponding start node information and the corresponding end node information in the input box. After acquiring the click and input information of the user, the client may generate input information of "start node information + end node information", attach the input information to the search request, and send the search request to the server. The server may perform the generation and presentation of the information stream based on the input information.

If the entity information can be extracted from the start node information and the end node information respectively, the entity information extracted from any one of the start node information or the end node information is preliminarily taken as the target entity information, and the other corresponding entity information is taken as the termination entity information. Then, the information flow of the target entity information may be extracted, and further extracting the information flow by taking each piece of associated entity information in the extracted information flow as the target entity information, and so on, if a piece of associated entity information is the terminating entity information, the associated entity information is not extracted any more. So that an information graph between the start node information and the end node information can be obtained.

For example, if the start node information is a material monomer, the end node information is a material application. By the method, the information flow can be gradually extracted from the material monomer until the material is applied, and the information map from the material monomer to the material application is obtained. Alternatively, the information flow can be gradually extracted from the material application to the material monomer, and an information map from the material monomer to the material application is obtained.

In still other embodiments, the method may further comprise: the server takes each entity information in the information flow as an information node interacting with the user, and performs visualization processing on the information flow by using a visualization method; sending the processed information flow to a client so that a user can check the information flow and trigger the information node through the client; and feeding back other information related to the information node to the client based on the triggering operation of the user on the information node. The information node is information corresponding to each entity information in the information flow. The trigger operation may include, for example, clicking, long pressing, sliding, and other operation modes. Other information associated with the inode may include, for example, a data source or other information flow not shown. The visualization method can adopt a Neo4j graphical method. As shown in fig. 14, fig. 14 shows an information graph diagram of an interactive visualization, where circles represent different information nodes, lines between the circles indicate that the information nodes are associated, and characters on the lines may represent association type description information between the information nodes, and may indicate association directions between the information nodes by arrows. In some embodiments, different colors may also be used to represent the data source from which the information came.

For example, a user may click on an inode to view the data source of the inode. Or, by sliding the information node to the left or right based on the information node, the non-displayed lower level information or the upper level information corresponding to the information node is viewed, where the lower level information may refer to information from the material corresponding to the current information node to the material application, and the upper level information may refer to information from the material corresponding to the current information node to the material monomer. Or sliding upwards or downwards by taking the information node as a reference to view the unexposed parameter entity information and supplier/manufacturer entity information corresponding to the information node. The parameter entity information can be information such as material type, process type, material attribute, unit or measurement. The display can be carried out in a mode of interaction with the user through design, and the use experience of the user can be further improved.

Based on the above application scenario, in other embodiments, the method may further include: the server extracts the target entity information and the parameter entity information of the associated entity information from a data source, wherein the parameter entity information at least comprises entity information corresponding to one of a material structure type entity, a process method entity, a material attribute entity, a unit entity or a measurement entity; and associating the extracted parameter entity information with corresponding target entity information and associated entity information. In some embodiments, different parametric entity information is extracted and associated with corresponding entity information, such as may be used with deep learning techniques.

Further detailed description information of the entity information is usually also present in the data source, such as description information of the process entity information, which may also include process method type, process parameters, manufacturer information, etc. For example, for a polymerization process, more specific description information of the polymerization process, such as specific process methods of solution polymerization, bulk polymerization, suspension polymerization, emulsion polymerization, etc., may exist in the data source, and at the same time, a great deal of information about the parameter conditions used in the polymerization, such as temperature, pressure, surrounding environment, production environment, etc., may be provided. The information can be further extracted, parameter entity information such as material attribute entities, unit entities or measurement entities of materials involved in the process method and the process or environmental parameters, unit entities, measurement entities and the like of the surrounding environment are associated with corresponding entity information, and a user can more accurately and comprehensively obtain specific parameter information.

The main attribute classes are shown in FIG. 15. Each property class has a sub-class of a different property. For example, the spectroscopy class can be divided into subcategories including raman spectroscopy, X-ray diffraction, and the like. FIG. 16 is a schematic diagram of linking material to a list of attributes from different attribute classes using units and metrics. FIG. 17 is a schematic illustration of the extraction of attributes, units, and metrics from a data source. FIG. 18 is a schematic illustration of linking extracted attributes, units, and measures into corresponding materials. Fig. 19, 20, and 21 show different expressions of attributes, units, and measures in a data source and information extraction diagrams. FIG. 22 is a schematic diagram of linking extracted attributes, units, and measures into corresponding materials.

By the above unit/measure and attribute extraction, different attributes associated with the material and changes in the attributes when subjected to different treatments can be visualized, enabling the user to select the best material and treatment to meet their needs. Meanwhile, information such as material attributes and units/measures is extracted and is associated to the information flow, association can be carried out between material application and material attributes, and application corresponding to the material is recommended based on the material attributes.

By extracting information flow by using the scheme provided by one or more embodiments, a user can accurately and comprehensively know the specific scheme of all details of the existing specific application, process method and the like of the material, and can know the patentees who disclose the application and process method through the associated data source, so that the infringement risk of the user can be effectively avoided.

As for a user performing a patent infringement analysis, the patent claims generally include descriptions of materials, and if a new material manufacturing process is claimed in the patent, the applicant generally needs to add all of the details of the manufacturing process in the claims, however, if the patent involves a modification to an existing manufacturing process, the applicant does not have to add all of the details of the manufacturing process. If an application-material link is newly discovered, the applicant can also add to the application of the material in the patent claims, however, the applicant usually prefers to describe a broader field of application in the patent claims, and more specific applications are not usually described in the claims. And descriptions of material properties, attributes and the like that are recited in the claims are often summarized to provide further scope of protection.

The lack of a detailed description of materials, manufacturing processes and applications in many patent claims makes accurate and comprehensive retrieval of much information difficult. If a third party makes or uses a material with some modification to the already patented base material, or with slight modification to the manufacturing process, but that modification may have been covered by the broad claims of other manufacturing patents, or the application for which the patent is being claimed falls within the application area of previously issued patents, and the third party fails to make a comprehensive and accurate search, the third party will likely face a higher risk of piracy.

Based on the methods provided by the foregoing embodiments, in still other embodiments of the present specification, an information flow extraction method is further provided, and is applied to a server, as shown in fig. 23, the method may include:

s40: acquiring target entity information;

s42: extracting at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information;

s44: and associating the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information.

In other embodiments, the associating the target entity information and the associated entity information by using the association relationship information may include:

In other embodiments, when the association relationship information includes peer description information, the association entity information corresponding to the peer description information is supplemented to the target entity information, and the target entity information is updated; the collocated entity description information comprises description information describing other expression forms of associated entity information as target entity information;

In still other embodiments, the method may further comprise:

In other embodiments, the extracting at least one associated entity information corresponding to the target entity information and the association relationship information between the target entity information and the associated entity information from the data source may include:

retrieving a data source where the target entity information is located;

locating context information of the target entity information in a data source;

In still other embodiments, the method may further comprise:

In still other embodiments, the method may further comprise: and associating the sub information flow of the target entity information with a corresponding data source.

In still other embodiments, the method may further comprise:

For the specific implementation of the information stream extraction method provided in the foregoing embodiment, reference may be made to the implementation of the foregoing method, and repeated details are not described here.

The information flow extraction device provided in one or more embodiments of the present specification may further verify and filter the extracted entity information by effectively using the association relationship information by simultaneously extracting the entity information and the association relationship information between the entity information, so as to determine whether the extracted entity information is entity information related to the target entity information, thereby further improving the accuracy of extracting the entity information related to the target entity information, and effectively filtering noise. Meanwhile, the extracted entity information can be associated with the target entity information by utilizing the association relationship information, the association relationship between the extracted entity information and the target entity information is effectively displayed, the user can conveniently check and comb the association relationship, the user can accurately and efficiently obtain the required useful information, the required or new solution is found, and the use experience of the user is improved.

Based on the information flow extraction method, one or more embodiments of the present specification further provide an information flow extraction apparatus. The apparatus may include systems, software (applications), modules, components, servers, etc. that utilize the methods described in the embodiments of the present specification in conjunction with hardware implementations as necessary. Since the implementation scheme of the apparatus for solving the problem is similar to that of the method, the specific implementation of the apparatus in the embodiment of the present specification may refer to the implementation of the foregoing method, and repeated details are not repeated. Specifically, fig. 24 is a schematic block diagram of an embodiment of an information flow extraction apparatus provided in the specification, and as shown in fig. 24, the apparatus may include:

a first obtaining module 402, configured to obtain target entity information;

a first extracting module 404, configured to extract, from a data source, at least one piece of associated entity information corresponding to the target entity information and associated relationship information between the target entity information and the associated entity information, where the associated relationship information includes information describing an association direction and an association type between entity information;

the first associating module 406 may be configured to associate the target entity information and the associated entity information by using the association relationship information, and obtain an information stream corresponding to the target entity information.

In other embodiments, the first association module 406 may include:

the first association unit may be configured to acquire at least one piece of associated entity information extracted from a data source and association relationship information between the target entity information and corresponding associated entity information, and associate the target entity information and corresponding associated entity information by using the association relationship information between the target entity information and corresponding associated entity information to obtain a sub information stream of the target entity information;

the second associating unit may be configured to link sub information streams obtained from the same or different data sources by using the target entity information as reference information, and obtain an information stream corresponding to the target entity information.

In some embodiments, when the association relationship information includes peer description information, the apparatus may further include:

the first extracting module 404 may be further configured to extract at least one associated entity information corresponding to the updated target entity information and association relationship information between the target entity information and the associated entity information from a data source.

In other embodiments, the apparatus may further comprise:

the second extraction module may be configured to extract information streams from the associated entity information extracted from the data source as target entity information, and obtain information streams corresponding to a plurality of target entity information;

the second association module may be configured to link information streams corresponding to the multiple pieces of target entity information with the corresponding pieces of target entity information as reference information, so as to obtain an information map.

In other embodiments, the first extraction module 404 may include:

the retrieval unit can be used for retrieving the data source where the target entity information is located;

the positioning unit can be used for positioning the context information of the target entity information in the data source;

the extracting unit may be configured to, when the context information includes a material process entity, a material name entity, and/or a material application entity, extract associated entity information corresponding to the target entity information and association relationship information between the target entity information and the associated entity information according to a material process information association manner.

In other embodiments, the apparatus may further comprise:

a third extraction module, which can be used for extracting the target entity information and the trade name of the associated entity information from a data source;

the third associating module may be configured to extract a manufacturer or a supplier corresponding to the target entity information or the associated entity information according to the trade name, and associate the manufacturer or the supplier with the target entity information or the associated entity information.

In other embodiments, the apparatus may further comprise:

the fourth correlation module may be configured to correlate the extracted parameter entity information with corresponding target entity information and correlated entity information.

In other embodiments, the first associating module 406 may further include:

a third associating unit, configured to associate the sub information stream of the target entity information with a corresponding data source.

In other embodiments, the apparatus may further comprise:

the visualization processing module can be used for taking each entity information in the information flow as an information node interacting with a user and carrying out visualization processing on the information flow by using a visualization method;

the first sending module may be configured to send the processed information stream to a client, so that a user views the information stream through the client and triggers the information node;

the second sending module may be configured to feed back, to the client, other information associated with the information node based on a trigger operation of the user on the information node.

It should be noted that the above-described apparatus may also include other embodiments according to the description of the method embodiment. The specific implementation manner may refer to the description of the related method embodiment, and is not described in detail herein.

The method or apparatus provided by the present specification and described in the foregoing embodiments may implement service logic through a computer program and record the service logic on a storage medium, where the storage medium may be read and executed by a computer, so as to implement the effect of the solution described in the embodiments of the present specification. Accordingly, the present specification also provides an information stream extraction apparatus comprising a processor and a memory storing processor-executable instructions which, when executed by the processor, implement steps comprising the method of any one of the above embodiments.

The storage medium may include a physical device for storing information, and typically, the information is digitized and then stored using an electrical, magnetic, or optical media. The storage medium may include: devices that store information using electrical energy, such as various types of memory, e.g., RAM, ROM, etc.; devices that store information using magnetic energy, such as hard disks, floppy disks, tapes, core memories, bubble memories, and usb disks; devices that store information optically, such as CDs or DVDs. Of course, there are other ways of storing media that can be read, such as quantum memory, graphene memory, and so forth.

It should be noted that the above description of the apparatus according to the method embodiment may also include other embodiments. The specific implementation manner may refer to the description of the related method embodiment, and is not described in detail herein.

The information flow extracting device according to the above embodiment extracts the entity information and the association relationship information between the entity information at the same time, and can further verify and filter the extracted entity information by effectively using the association relationship information to determine whether the extracted entity information is the entity information related to the target entity information, thereby further improving the accuracy of extracting the entity information related to the target entity information and effectively filtering noise. Meanwhile, the extracted entity information can be associated with the target entity information by utilizing the association relationship information, the association relationship between the extracted entity information and the target entity information is effectively displayed, the user can conveniently check and comb the association relationship, the user can accurately and efficiently obtain the required useful information, the required or new solution is found, and the use experience of the user is improved.

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides an information flow display method, which is applied to a server, where the method may include:

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides a server, which may include:

the first receiving module may be configured to receive an information stream acquisition request sent by a client, where the information stream acquisition request includes input information acquired by the client;

a fifth extraction module, configured to extract one or more pieces of entity information from the input information as target entity information;

a third sending module, configured to send the information stream to the client for displaying on the client.

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides an information flow display method, which is applied to a client and may include:

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides a client, which may include:

the fourth sending module may be configured to send an information flow obtaining request to the server, where the information flow obtaining request includes input information obtained by the client; enabling the server to receive the information flow obtaining request and extract one or more entity information from the input information to serve as target entity information; extracting at least one piece of associated entity information corresponding to the target entity information and associated relation information between the target entity information and the associated entity information from a data source, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information; associating the target entity information and the associated entity information by using the association relation information to obtain an information stream corresponding to the target entity information, and sending the information stream to the client;

the second receiving module can be used for receiving the information stream sent by the server;

a first presentation module may be configured to present the information stream.

The method or apparatus provided by the present specification and described in the foregoing embodiments may implement service logic through a computer program and record the service logic on a storage medium, where the storage medium may be read and executed by a computer, so as to implement the effect of the solution described in the embodiments of the present specification. Accordingly, the present specification also provides an information flow presentation device comprising a processor and a memory storing processor-executable instructions which, when executed by the processor, implement steps comprising the method of any one of the above embodiments.

In other implementation scenarios, the user also usually uses an information retrieval method to query the data source or the patentee behind the data source when performing analysis such as technical query and infringement analysis. However, generally, information may have a plurality of different description forms in the same data source or different data sources, and even a large amount of incomplete and non-descriptive information exists, so that the searched information has the problems of inaccuracy and incompleteness. And the user needs to expend great energy to screen and comb when checking, and the experience is poor. Accordingly, as shown in fig. 25, in other embodiments of the present specification, an information retrieval method may further be provided, where the method may include:

s60: a client sends a retrieval request, wherein the retrieval request comprises input information acquired by the client;

s62: the server receives a retrieval request sent by a client, and extracts entity information from the input information, wherein the entity information comprises an entity type and an entity value;

s64: the server determines a data source where the entity information is located according to associated entity information of the entity information in the data source and associated relation information between the entity information and the associated entity information, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information;

s66: and the server sends the retrieval result to the client so as to display the client.

The input information can be reference information of information retrieval, and the server retrieves a data source directly describing or indirectly describing the input information by taking the input information as a reference. The associated entity information of the entity information in the data source and the association relationship information between the entity information and the associated entity information may be implemented by referring to the embodiments of the information stream extraction method, which are not described herein again.

Figure 26 shows a schematic of the search interface and search results using the keyword search for "fullerenes as polymer additives". The darker the gray scale shown in fig. 26 is the non-correlation search and the weaker the correlation search result. As shown in fig. 26, only 3 related patents are obtained by keyword search. Fig. 27 is a schematic view of a search interface according to an embodiment of the present disclosure, which is used to perform information search based on a material information flow. Fig. 28 is a schematic diagram showing a retrieval result of the scenario of the embodiment of the present specification, and all relevant searches are accurately extracted by the scenario of the embodiment of the present specification.

There may be no specific explicit expression of "Fullerene as a polymer additive" in many data sources, or even if there is an expression of "Fullerene as a polymer additive", the expression may not be in the same form as the expression input by the user, resulting in many data sources that cannot be retrieved efficiently by way of keyword matching. Alternatively, when the user inputs the keyword, the user may use an or method to separate different keyword information, but in this way, more noise information may be retrieved.

By using the scheme of the embodiment of the present specification, by searching the data source for the associated entity information of the entity information in the input information and the association relationship information between the entity information and the associated entity information, even if the data source does not explicitly or specifically have "Fullerene" as the description information of the "additive", the entity information can still be used: the additive entity-Fullerene accurately positions a data source describing the entity information through the associated entity information and the associated relation information between the associated entity information and the associated entity information, and further effectively extracts all data sources describing that the 'Fullerene' is used as an additive role, so that signals (related search) can be effectively separated from noise (unrelated search), and the search accuracy is greatly improved. Further limiting the condition of "polymer" can eliminate the data source describing that "Fullerene" is used as an "additive" in the information stream corresponding to other material types, thereby further improving the accuracy of information retrieval.

In some embodiments, a name variant of "fullerrene", i.e., parity information, may be further extracted, such as the chinese description "Fullerene". By further updating the parity information into the reference information corresponding to the search, the accuracy and comprehensiveness of the search result can be further improved.

In some embodiments, classification retrieval can be carried out according to material structures or material applications in information retrieval. Such as may be added to the input information by material structure or material application at the time of information retrieval.

The material structure may refer to the combination or arrangement of atoms (or ions, molecules) or constituent elements (i.e., structural elements) of the constituent materials and the combination, arrangement and various relationships of the structural elements in a certain order. Different materials have various structural elements, for example, various phases, tissues, defects, monomers, macromolecular chains, and the like of the materials belong to the structural elements of the materials. The materials are classified according to their structure, such as the material classes that may include polymers, metals, ceramics, biological, composites, etc., and for each of these classes, further more refined classes may be provided, as shown in fig. 31. FIG. 31 shows a schematic diagram of the classification of polymers into further refinement of their material structure.

The material application can comprise the application field or specific application corresponding to the material. Application area classifications may include, for example, construction materials, energy materials, printed materials, optoelectronic materials, and the like. The division into different fields of application can be further divided up to specific applications, such as organic electro-display coatings in electro-optical materials. As shown in fig. 33, fig. 33 shows a further refined classification diagram of smart materials.

Entity types corresponding to materials with different material structures or incidence relation information description forms among the entity information, and information flow characteristics from a material monomer to material application may have great difference, and by classifying the material structures in advance and searching based on the material structure categories to which the materials to be searched, the material application or the material process belong, the accuracy of extracting the incidence relation information among the entity information and the entity information can be further improved, so that the searched data source is more in line with the requirements of users.

Corresponding to different application fields or different specific applications, the related raw materials, processing technology, manufacturing method and the like may have larger differences, and by further refining the material application and extracting the entity information and the incidence relation information between the entity information based on the material application, the accuracy of information extraction can be further improved, noise interference is reduced, and the retrieved data source is more in line with the requirements of users. Meanwhile, for some data sources, such as patents, papers, etc., many retrieval systems can also be classified according to the application fields, as shown in fig. 29. The left graph of fig. 29 represents a larger application classification, and the right graph represents an application classification in which a certain application classification in the left graph is further specified. Therefore, by considering the material application during the retrieval, the data source corresponding to the corresponding application field can be retrieved in a targeted manner, and the retrieval efficiency and the retrieval accuracy are greatly improved.

Particularly for the retrieval of image information, the difficulty of detecting a material graph on a fine granularity level can be reduced by carrying out classification retrieval according to the material structure and the material application, different materials and material patterns can be detected from the image, and the retrieval accuracy is improved.

For example, as shown in fig. 34 to 36, fig. 34 is a schematic diagram illustrating a search result searched by using a keyword; FIG. 35 is a schematic diagram of a search interface provided by embodiments of the present disclosure; fig. 36 is a schematic diagram showing a search result using the above scheme provided in the embodiment of the present specification. As can be seen from fig. 36, in the embodiments of the present specification, by further limiting the application field during the retrieval, the server further considers the patent to which the material is applied, and then extracts the entity information and the association relationship information between the entity information, so as to further improve the accuracy of the results of the image search, the image extraction, the image object detection, and the image classifier.

For example, as shown in fig. 37 and 38, fig. 37 shows an image schematic diagram of a protein structure, a DNA plasmid, and a material microstructure, and fig. 38 shows an image schematic diagram of a circuit diagram and a flowchart. When the protein structure, the DNA plasmid and the material microstructure are searched, patents corresponding to noise images with specific image types (such as circuit diagrams, flow charts and the like) can be effectively eliminated by considering the material structure classification, and the images of the protein structure, the DNA plasmid and the material microstructure can be accurately extracted.

Based on the above application scenario example, in some embodiments, the information input interface presented by the client may further include an information input area, a first selection list, and/or a second selection list. The information input area can be used for information input by a user, and the first selection list and the second selection list can be used for information selection by the user. The first selection list may include category information of material applications or material structures. The second selection list may include category information of entity types. Correspondingly, the client side obtains input information based on the information input area, the first selection list and/or the second selection list.

As shown in fig. 27, the user can input information in the input box and limit the entity type of the input information to perform information retrieval, and by performing information retrieval in this way, the entity type and the entity value can be accurately extracted from the input retrieval information, and the accuracy of information retrieval can be improved.

As shown in fig. 27 and fig. 35, application fields or structure categories may also be defined, and by defining application fields or structure categories, extraction features of entity information corresponding to the corresponding application fields or structure categories and association relationship information between the entity information may be determined conveniently, so that accuracy of information extraction is improved. Meanwhile, the input information can be more normalized by a preset material structure and a material application classification selection mode, and the accuracy of information retrieval is improved.

In some embodiments, the first selection list or the second selection list may further display information to be selected in an interactive visualization format. The interactive visualization format shows a list selection format for displaying entity types or material structures and category information of material applications in a visualization form and determining corresponding categories and sub-categories by receiving a trigger operation for each category information, as shown in fig. 30 to 33.

FIG. 30 shows several larger category presentations of material structures that a user may click on, and the client may determine that the user selected a material structure as part of the input information based on the user's clicking action. The client may further expose several larger categories. The user may further click on any of the categories, such as clicking on a polymer. The client may determine that the user further selected a polymer as part of the input information based on a click operation of the user. The client may further present a refined classification of the polymer, as shown in fig. 31. The user may further click on any of the categories in fig. 31, and the client may determine a detailed category further selected by the user as part of the input information based on the user's click operation. Of course, the user may operate the device in other manners. Fig. 32 shows several larger category displays for material applications, and fig. 33 shows a detailed classification diagram for smart materials. The triggering and displaying embodiments for the material application may be of the same material structure, and are not described herein.

By adopting the interactive visualization format as the first selection list or the second selection list, the material structure and the material application detailed and classified display can be facilitated, meanwhile, the interactivity can be improved, and the user experience is improved.

In other embodiments, the material application may also be linked to material properties and exhibit properties related to a particular application, as shown in FIG. 39, where FIG. 39 exhibits 10 large properties of smart materials. The internal structure of the material can be changed along with the change of chemical compositions and external conditions, thereby changing the performance of the material. For example, low carbon steels with carbon fractions below 0.25% by mass generally have good ductility and toughness, but low strength and hardness; high carbon steel with carbon mass fraction in the range of 0.6-1.4% has high strength and hardness, and poor plasticity and toughness. Different properties of the material have a great influence on the practical application of the material, such as materials with different yield strengths, and the application fields thereof have great difference. By establishing the association between the material application and the material attribute and describing different units/measurement values of the material attribute in the data source, the extraction of the material application information can be effectively determined and verified, and the accuracy of information extraction and retrieval is improved.

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides an information retrieval method, which is applied to a server and may include:

and sending the retrieval result to the client to display the client.

the third receiving module may be configured to receive a retrieval request sent by a client, where the retrieval request includes input information acquired by the client;

the data source determining module may be configured to determine a data source where the entity information is located according to associated entity information of the entity information in the data source and association relationship information between the entity information and the associated entity information, where the association relationship information includes information describing an association direction and an association type between the entity information;

a fifth sending module, configured to send the search result to the client, so that the client performs presentation.

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides an information retrieval method, which is applied to a client and may include:

and receiving and displaying the retrieval result sent by the server.

In other embodiments, the information input interface presented by the client may include an information input area, a first selection list, and/or a second selection list; the information input area can be used for inputting information; the first selection list and the second selection list can be used for information selection, and the first selection list can comprise category information of material application or material structure; the second selection list comprises category information of entity types;

accordingly, the client may obtain input information based on the information input area, the first selection list, and/or the second selection list.

In other embodiments, the first selection list and/or the second selection list may display the information to be selected in an interactive visualization format; the interactive visualization format may represent a list selection format for displaying entity types or material structures, category information of material applications in a visualization form, and determining corresponding categories and subcategories by receiving a trigger operation on each category information.

a sixth sending module, configured to send a retrieval request to a server, where the retrieval request includes input information obtained by a client; so that the server extracts entity information from the input information, wherein the entity information comprises an entity type and an entity value; determining a data source where the entity information is located according to associated entity information of the entity information in the data source and associated relation information between the entity information and the associated entity information, wherein the associated relation information comprises information describing an associated direction and an associated type between the entity information; sending the retrieval result to the client;

a fourth receiving module, configured to receive the retrieval result sent by the server;

and the second display module can be used for displaying the retrieval result.

In other embodiments, the client may further include:

the information input interface display module can be used for displaying an information input interface, and the information input interface can comprise an information input area, a first selection list and/or a second selection list; the information input area is used for inputting information; the first selection list and the second selection list can be used for information selection, and the first selection list can comprise category information of material application or material structure; the second selection list can comprise category information of entity types;

an input information obtaining module may be configured to obtain input information based on the information input area, the first selection list, and/or the second selection list.

The method or apparatus provided by the present specification and described in the foregoing embodiments may implement service logic through a computer program and record the service logic on a storage medium, where the storage medium may be read and executed by a computer, so as to implement the effect of the solution described in the embodiments of the present specification. Accordingly, the present specification also provides an information retrieval device comprising a processor and a memory storing processor-executable instructions which, when executed by the processor, implement steps comprising the method of any one of the above embodiments.

In general, a patent or paper document may describe a great deal of information, and the abstract information given by the author or applicant does not effectively represent the main information of the whole patent or paper document. When technical inquiry is carried out, users are often used to preliminarily determine whether the required information exists in the current document only by abstracting due to the increase of the information quantity, so that the information is missed. However, if overlooking is avoided, the user is required to expend a great deal of effort and to review and analyze the entire contents of the document, which is time consuming and labor intensive. Accordingly, as shown in fig. 40, in other embodiments of the present specification, a summary information generating method may further be provided, where the method may include:

s80: a client sends a summary information generation request, wherein the summary information generation request comprises a data source of a summary to be generated;

s82: the server receives the summary information generation request sent by the client;

s84: the server extracts entity information and incidence relation information between the entity information from the data source, and associates the corresponding entity information according to the extracted incidence relation information; wherein, the incidence relation information comprises information describing the incidence direction and the incidence type between the entity information;

s86: and the server generates abstract information according to the associated entity information and sends the generated abstract information to the client so as to display the client.

The data source may include patent application text, papers, or other types of information text for which summary information is to be generated. After the server obtains the corresponding information text, the server can use NLP to process and label the original information text. After marking, entity information in the information text can be extracted by using NER based on a pre-constructed entity information dictionary or an entity information base, and a material name entity or a material application entity is extracted. And using the extracted entity information as an object to be analyzed.

Then, the server may acquire metadata of the object to be analyzed using the NLP algorithm. For example, entity information in the context information of the object to be analyzed may be extracted, and the interference word may be deleted using a filtering algorithm. The interfering words include stop words, common words, non-scientific words and words that are not related to the material context. The filtered entity information may then be attached to these objects as metadata.

Then, the metadata of the objects to be analyzed can be compared to establish an association relationship between the two objects to be analyzed. Then, the association relationship may be utilized to associate the objects to be analyzed. Therefore, the incidence relation between important entity information in the information text is established, an information abstract is formed, and the information abstract is fed back to the client side for displaying. The expression form of the information abstract can be a segment of text information generated based on the entity information and the incidence relation information between the entity information, and can also be an information stream formed based on the entity information and the incidence relation information between the entity information.

For a plurality of information texts, the entity information and the association relation information extracted from the two information texts can be compared with each other, and the similarity between the entity information and the association relation information between the two information texts can be found. Entity similarity can be considered by comparing multiple attributes related to different contexts and metadata. Each entity may have a unique threshold for entity matching. Entities and relationships are grouped together to form important workflow links between two documents, such as where similarities/matches can be found between entities based on entity thresholds.

The extraction of the entity information and the association relationship information between the entity information may be implemented by referring to the information flow extraction method, which is not described herein again. By generating the summary information in the manner, the entity information in the data source information and the incidence relation between the entity information can be accurately and comprehensively extracted, screening and combing of users are avoided, and the users can find the required information more quickly and accurately.

The generated summary information can be displayed in the form of characters, tables and the like, and can also be displayed in the form of information flow. The corresponding information flow extraction method can be performed by referring to the above embodiment, and the summary information is displayed in an information flow manner, so that the user can check the information more conveniently, and the user experience is improved.

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides a summary information generating method, which is applied to a server and may include:

the fifth receiving module may be configured to receive a summary information generation request sent by a client, where the summary information generation request includes a data source to be summarized;

the eighth extraction module may be configured to extract entity information and association relationship information between the entity information from the data source; wherein, the incidence relation information comprises information describing the incidence direction and the incidence type between the entity information;

the sixth association module may be configured to associate the corresponding entity information according to the extracted association relationship information;

the seventh sending module may be configured to send the generated summary information to the client, so that the client performs presentation.

Based on the method provided by the foregoing embodiment, an embodiment of this specification further provides a summary information generating method, which is applied to a client, and may include:

and receiving and displaying the summary information sent by the server.

the eighth sending module may be configured to send a summary information generation request to the server, where the summary information generation request includes a data source to be summarized; so that the server extracts the entity information and the incidence relation information between the entity information from the data source, and correlates the corresponding entity information according to the extracted incidence relation information; wherein, the incidence relation information comprises information describing the incidence direction and the incidence type between the entity information; generating abstract information according to the associated entity information, and sending the generated abstract information to the client;

a fifth receiving module, configured to receive the summary information sent by the server;

and the third display module can be used for displaying the summary information.

The method or apparatus provided by the present specification and described in the foregoing embodiments may implement service logic through a computer program and record the service logic on a storage medium, where the storage medium may be read and executed by a computer, so as to implement the effect of the solution described in the embodiments of the present specification. Accordingly, the present specification also provides a summary information generating device, comprising a processor and a memory storing processor-executable instructions, which when executed by the processor implement the steps of the method according to any one of the above embodiments.

The present specification also provides a system, which may be a single information flow extraction system, or an information flow presentation system, or an information retrieval system, or a summary generation system, and may also be applied to a plurality of information extraction systems. The system may be a single server, or may include a server cluster, a system (including a distributed system), software (applications), an actual operating device, a logic gate device, a quantum computer, etc. using one or more of the methods or one or more of the example devices of the present specification, in combination with a terminal device implementing hardware as necessary. The information retrieval system may include at least one processor and a memory storing computer-executable instructions that, when executed by the processor, implement the steps of the method described in any one or more of the embodiments above.

It should be noted that the above-mentioned system may also include other implementation manners according to the description of the method or apparatus embodiment, and specific implementation manners may refer to the description of the related method embodiment, which is not described in detail herein.

The embodiments of the present description are not limited to what must be consistent with a standard data model/template or described in the embodiments of the present description. Certain industry standards, or implementations modified slightly from those described using custom modes or examples, may also achieve the same, equivalent, or similar, or other, contemplated implementations of the above-described examples. The embodiments using these modified or transformed data acquisition, storage, judgment, processing, etc. may still fall within the scope of the alternative embodiments of the present description.

The description has been presented with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the description. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method or apparatus that comprises the element.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment. In the description of the specification, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the specification. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.

The above description is only an example of the present specification, and is not intended to limit the present specification. Various modifications and alterations to this description will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.

Claims

1. An information flow presentation method, comprising:

2. An information flow extraction method, comprising:

acquiring target entity information;

3. The method according to claim 2, wherein the associating the target entity information and the associated entity information by using the association relationship information comprises:

4. The method according to claim 2, wherein when the association relationship information includes peer description information, the associated entity information corresponding to the peer description information is supplemented to the target entity information, and the target entity information is updated; the collocated entity description information comprises description information describing other expression forms of associated entity information as target entity information;

5. The method of claim 2, further comprising:

6. The method according to claim 2, wherein the extracting at least one associated entity information corresponding to the target entity information and the association relationship information between the target entity information and the associated entity information from the data source includes:

retrieving a data source where the target entity information is located;

locating context information of the target entity information in a data source;

7. The method of claim 2, further comprising:

8. The method of claim 2, further comprising:

9. The method of claim 3, further comprising: and associating the sub information flow of the target entity information with a corresponding data source.

10. The method of claim 2, further comprising:

11. An information stream extraction apparatus, comprising:

the first acquisition module is used for acquiring target entity information;

12. The apparatus of claim 11, wherein the first associating module comprises:

13. The apparatus according to claim 11, wherein when the association information includes peer description information, the apparatus further comprises:

14. The apparatus of claim 11, further comprising:

15. The apparatus of claim 11, wherein the first extraction module comprises:

16. The apparatus of claim 11, further comprising:

17. The apparatus of claim 11, further comprising:

18. The apparatus of claim 12, wherein the first associating module further comprises:

19. The apparatus of claim 11, further comprising:

20. An information stream extraction device, characterized in that the device comprises a processor and a memory for storing processor-executable instructions, which when executed by the processor implement the steps of the method of any of the preceding claims 2-10.

21. An information flow presentation method applied to a server includes:

22. A server, comprising:

23. An information flow presentation method applied to a client includes:

24. A client, comprising:

and the first display module is used for displaying the information flow.

25. An information retrieval method, applied to a server, includes:

and sending the retrieval result to the client to display the client.

26. A server, comprising:

27. An information retrieval method is applied to a client and comprises the following steps:

and receiving and displaying the retrieval result sent by the server.

28. The method of claim 27, wherein the information input interface presented by the client comprises an information input area, a first selection list and/or a second selection list; the information input area is used for inputting information; the first selection list and the second selection list are used for information selection, and the first selection list comprises category information of material application or material structures; the second selection list comprises category information of entity types;

29. The method according to claim 28, wherein the first selection list and/or the second selection list presents information to be selected in a format comprising interactive visualization; the interactive visualization format shows entity types or material structures and category information of material applications in a visualization form, and determines a list selection format of corresponding categories and subcategories by receiving trigger operations on the category information.

30. A client, comprising:

and the second display module is used for displaying the retrieval result.

31. The client of claim 30, further comprising:

32. A summary information generation method is applied to a server and comprises the following steps:

33. A server, comprising:

34. A summary information generation method is applied to a client and comprises the following steps:

and receiving and displaying the summary information sent by the server.

35. A client, comprising:

and the third display module is used for displaying the summary information.