WO2019047352A1 - 基于社交数据的资产配置方法、电子装置及介质 - Google Patents

基于社交数据的资产配置方法、电子装置及介质 Download PDF

Info

Publication number
WO2019047352A1
WO2019047352A1 PCT/CN2017/108795 CN2017108795W WO2019047352A1 WO 2019047352 A1 WO2019047352 A1 WO 2019047352A1 CN 2017108795 W CN2017108795 W CN 2017108795W WO 2019047352 A1 WO2019047352 A1 WO 2019047352A1
Authority
WO
WIPO (PCT)
Prior art keywords
preset
social data
asset
investor
subjective
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2017/108795
Other languages
English (en)
French (fr)
Inventor
毕野
肖京
王建明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2019047352A1 publication Critical patent/WO2019047352A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/284—Lexical analysis, e.g. tokenisation or collocates
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/06—Asset management; Financial planning or analysis

Definitions

  • the present application relates to the field of computer technologies, and in particular, to a social data-based asset configuration method, an electronic device, and a medium.
  • the existing big data asset allocation model is mainly based on the fundamental analysis of different assets and the analysis of technical indicators to design effective factors, construct a quantitative asset model, predict the risks and benefits of different types of assets, and finally formulate the portfolio of suitable investors.
  • the existing asset allocation model can only be based on objective data (such as fundamental analysis, technical indicator analysis) for asset allocation, without considering the subjective influence factor of investors, and there is a lack of timeliness and comprehensiveness.
  • the purpose of the present application is to provide a social data-based asset configuration method, an electronic device, and a medium, which are intended to improve the timeliness of asset allocation.
  • a first aspect of the present application provides an electronic device including a memory, a processor, and a social data-based asset configuration system stored on the memory and operable on the processor,
  • the social data based asset configuration system is implemented by the processor to implement the following steps:
  • A. Obtain social data related to the preset category asset in the latest preset time period from the preset data source;
  • the second aspect of the present application provides a social data-based asset configuration method, where the social data-based asset configuration method includes:
  • A. Obtain social data related to the preset category asset in the latest preset time period from the preset data source;
  • a third aspect of the present application provides a computer readable storage medium storing a social data based asset configuration system, the social data based asset configuration system being executable by at least one processor to enable The at least one processor performs the steps of the social data based asset configuration method as described above.
  • the social data-based asset allocation method, the electronic device and the medium proposed by the application determine the emotional analysis and subjective expectation of the overall investor on the financial investment product by analyzing the social data of the latest preset time period of the investor, and based on the investment
  • the above subjective influence factors are used to perform asset allocation.
  • FIG. 1 is a schematic diagram of an operating environment of a preferred embodiment of an asset allocation system 10 of the present application
  • FIG. 2 is a schematic flowchart of an embodiment of a social data-based asset configuration method according to an embodiment of the present application
  • FIG. 3 is a schematic diagram of the refinement process of step S50 shown in FIG. 2.
  • first, second and the like in the present application are for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. .
  • features defining “first” and “second” may include at least one of the features, either explicitly or implicitly.
  • the technical solutions between the various embodiments may be combined with each other, but must be based on the realization of those skilled in the art, and when the combination of the technical solutions is contradictory or impossible to implement, it should be considered that the combination of the technical solutions does not exist. Nor is it within the scope of protection required by this application.
  • FIG. 1 is a schematic diagram of an operating environment of a preferred embodiment of the asset configuration system 10 of the present application.
  • the asset configuration system 10 is installed and operated in the electronic device 1.
  • the electronic device 1 may include, but is not limited to, a memory 11, a processor 12, and a display 13.
  • Figure 1 shows only the electronic device 1 with components 11-13, but it should be understood that not all illustrated components may be implemented, and more or fewer components may be implemented instead.
  • the memory 11 comprises at least one type of readable storage medium, which in some embodiments may be an internal storage unit of the electronic device 1, such as a hard disk or memory of the electronic device 1.
  • the memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in hard disk equipped on the electronic device 1, a smart memory card (SMC), and a secure digital device. (Secure Digital, SD) card, flash card, etc.
  • SMC smart memory card
  • secure digital device Secure Digital, SD
  • the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device.
  • the memory 11 is configured to store application software and various types of data installed in the electronic device 1, such as program codes of the asset configuration system 10.
  • the memory 11 can also be used to temporarily store data that has been output or is about to be output.
  • the processor 12 in some embodiments, may be a central processing unit (CPU), a microprocessor or other data processing chip for running program code or processing data stored in the memory 11, for example The asset configuration system 10 and the like are executed.
  • CPU central processing unit
  • microprocessor or other data processing chip for running program code or processing data stored in the memory 11, for example
  • the asset configuration system 10 and the like are executed.
  • the display 13 in some embodiments may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch sensor, or the like.
  • the display 13 is used to display information processed in the electronic device 1 and a user interface for displaying visualization, such as the determined investor's latest subjective sentiment prediction viewpoint, the ratio of the assets of different categories in the asset configuration, and the like.
  • the components 11-13 of the electronic device 1 communicate with one another via a system bus.
  • the social data based asset configuration system 10 includes at least one computer readable instructions stored in the memory 11, the at least one computer readable instructions being executable by the processor 12 to implement various embodiments of the present application.
  • the social data-based asset configuration system 10 described above is executed by the processor 12 to implement the following steps:
  • Step S1 Obtain social data related to the preset category asset in the latest preset time period from the preset data source.
  • the social data related to the preset category asset in the latest preset time period of the investor is obtained from the preset data source, wherein the preset category asset is large.
  • assets as equity stocks, fixed investment bonds, alternative investment gold, derivatives option futures, etc.
  • the social data includes a comment or analysis text information of an investor in a preset range for a preset category asset within a preset preset time period, for example, a review of the investor's latest preset time period for different categories of assets, Analyze and other textual information, based on the LDA subject classification method, the investor's comments on different major assets (stocks, bonds, gold, options, futures, etc.) are classified and collected.
  • the social data of the investor's latest preset time period may also be obtained from the investor's QQ, Weibo, WeChat, Snowball, Oriental Fortune and other social software, including but not limited to Investors actively send out articles and circle of friends related to the pre-defined category assets, as well as comments from investors on the content posted by others, forwarding content, and so on.
  • the social data of the financial type can be increased when the social data of the latest preset time period of the investor is obtained, such as increasing the social software from the investor financial system (such as snowball, Financial social data obtained on Oriental Wealth, etc., and the amount or weight of financial social data related to financial issues issued by investors.
  • Step S2 calculating and converting a plurality of vocabulary vectors based on the social data.
  • the social data may be segmented by a preset word segmentation method.
  • the word segmentation method of the string matching may be used to perform word segmentation processing on the social data, such as a forward maximum matching method, and the character string in one information is segmented from left to right, that is, the social data is left to right.
  • Several consecutive characters in the table match the vocabulary, if the match, a word is segmented; or, the reverse maximum matching method, the character string in a message is segmented from right to left, that is, from the social data
  • the end of the matching scan starts, and several consecutive characters in the information text of the word to be distinguished are matched with the vocabulary from right to left.
  • Word lexical segmentation can also be used to classify each piece of information. Word lexical segmentation is a segmentation method for machine speech judgment. It uses syntactic information and semantic information to deal with ambiguity phenomena to segment words. You can also use the statistical segmentation method to process each word segmentation, from the current user's historical search record or the history of the public user.
  • word segmentation can also be performed based on the Chinese word segmentation tool NLPIR, which is not limited herein.
  • the word vector model word2vec is used to convert each word segmentation into a vocabulary vector, wherein word2vec is a tool for converting words into a vector form, which can simplify the processing of the text content into a vector space.
  • the vector operation calculates the similarity in the vector space to represent the semantic similarity of the text.
  • word2vec can be used to simplify the processing of text content into vector operations in K-dimensional vector space, and the similarity in vector space can be used to represent the semantic similarity of text. Therefore, the word vector output by word2vec can be used to do things like clustering, finding synonyms, part of speech analysis, etc. Moreover, word2vec is very efficient.
  • Step S3 performing similarity calculation on each vocabulary vector and the preset keyword respectively, and determining that the keyword with the highest similarity with each vocabulary vector is the corresponding keyword of the vocabulary vector, and the keywords are pre-marked differently.
  • Subjective emotions predict the point of view.
  • a topic dictionary that characterizes different subjective emotion prediction opinions (long, short, etc.) is constructed, and keywords in the dictionary are marked according to different subjective emotion prediction viewpoints.
  • the different subjective sentiment prediction views of the mark include long-term, short-term forecasting opinions and the like.
  • keywords such as “promising”, “strong momentum” and “seeing more” in the dictionary can be marked as “long” predictions; the key words in the dictionary are “not optimistic”, “lower”, “bearish” and so on. Words can be marked as "short” predictions.
  • the converted lexical vector may be similarly calculated with the keywords in the dictionary marked with different subjective emotion prediction views, for example, Word2vec converts the keywords in the dictionary marked with different subjective emotion prediction into vector form, so that the similarity calculation of several lexical vectors and keywords can be simplified to the vector operation in vector space, and the similarity in vector space can be calculated.
  • Degree can be used to represent the semantic similarity between several lexical vectors and keywords.
  • the similarity calculation may be performed separately for each of the keywords in the keywords whose vocabulary vectors have different subjective emotion prediction views. For example, if the subjective sentiment forecasting perspective in the subjective sentiment predictive perspective dictionary includes “long” and “short” predictive opinions, the keywords that mark the "long” predictive perspective include “optimistic", “see more”, and mark “short” prediction views. Keywords include “lower” and “bearish”. After the user's social data is segmented and converted into several vocabulary vectors, each vocabulary vector of the converted vocabulary vectors can be similarly calculated with the keywords to determine the keywords with the highest similarity with each vocabulary vector. Is the corresponding keyword of the vocabulary vector.
  • the vocabulary vector A can be similarly calculated for each of the keywords of "optimistic”, “too much”, “lower”, and “bearish”.
  • the similarity between the vocabulary vector A and each keyword can be calculated, and the keyword with the highest similarity to the vocabulary vector A can be selected as the lexical direction.
  • the corresponding keywords of quantity A such as vocabulary vector A and the similarity of "optimistic”, “seeing more”, “lower” and “bearish” are 90%, 80%, 40%, 30%, respectively, and vocabulary vector A
  • the keyword with the highest similarity is “optimistic”, and “seeing good” is used as the corresponding keyword of vocabulary vector A.
  • step S4 all vocabulary vectors are sorted according to the similarity from high to low according to the similar keyword similarity, and a preset number of vocabulary vectors with the highest ranking are selected.
  • the similarity between the vocabulary vector a and the corresponding keyword "good” is 99%
  • the similarity between the vocabulary vector b and the corresponding keyword "see more” is 98%
  • the vocabulary vector c is similar to the corresponding keyword "lower”.
  • the degree is 97%
  • the similarity between the vocabulary vector d and the corresponding keyword "bearish” is 96%
  • the order of similarity from high to low is a, b, c, d.
  • a preset number for example, 50
  • Step S5 Acquire a corresponding keyword of the preset number of vocabulary vectors, and determine a user's latest subjective mood prediction viewpoint according to the subjective sentiment prediction viewpoint marked by the corresponding keyword.
  • the vocabulary vector with the highest similarity to the keywords in the dictionary such as TOP50
  • the corresponding keywords according to the selected TOP50 vocabulary vector may be selected.
  • the subjective sentiment is predicted to predict the investor's latest subjective sentiment predictions. For example, the number or proportion of long-term, short-term prediction opinions in the subjective emotion prediction viewpoint marked by the corresponding keyword of the selected TOP50 vocabulary vector can be counted, and the latest subjective emotion as the investor is selected in the selected quantity or the proportion Forecasting perspectives.
  • the number of long-predicted opinions in the subjective sentiment prediction point marked by the selected TOP50 vocabulary vector is 30, and the number of short-predicted opinions is 20, it means that the overall investor has a long-term forecasting viewpoint. For the majority, it is determined that the investor's latest subjective sentiment predictions are long-term predictions.
  • step S6 the proportion of the preset category assets in the preset asset allocation is adjusted based on the investor's latest subjective emotion prediction viewpoint.
  • the present embodiment adjusts preset asset types in the preset asset allocation based on the investor's latest subjective emotion prediction viewpoint.
  • preset asset types such as the proportion of large categories of assets. For example, if the investor's latest subjective sentiment forecasting view is a long-term forecasting viewpoint, increase the proportion of the pre-set asset allocation in the preset asset allocation; if the investor's latest subjective sentiment forecasting viewpoint is a short-term forecasting viewpoint, the pre-reduction is reduced.
  • real-time dynamic social data is introduced, and the investor's subjective tendency prediction for different assets is more accurate and real-time.
  • objective financial analysis basic analysis, technology
  • Indicator analysis while incorporating the overall investor's subjective emotional views and forecasting views, effectively improve the timeliness of asset allocation.
  • the present embodiment determines the emotional analysis and subjective expectation of the overall investor on the financial investment product by analyzing the social data of the latest preset time period of the investor, and based on the above-mentioned subjective influence factor of the investor. Perform asset allocation.
  • the method specifically includes:
  • investors corresponding to the social data are classified into common experience groups and rich experience groups according to a preset classification rule; for example, investors with high investment returns and/or investment experience may be classified based on historical investment data of investors To the extensive experience group, investors with low investment returns and/or low investment experience are divided into the general experience group.
  • the similarity calculation is performed on a plurality of vocabulary vectors and keywords whose preset subjective emotion prediction views are different, and the vocabulary vectors with the highest similarity among the vocabulary vectors corresponding to the common experience group and the rich experience group are respectively selected. For example, the similarity calculations are compared between the different grouped vocabulary vectors and the keywords marked with different subjective emotion prediction views in the dictionary, and the vocabulary list with the highest similarity of the different experience groups TOP1000 is selected.
  • the subjective sentiment prediction views of the common experience group and the rich experience group are respectively analyzed according to the preset analysis rules, and the subjective emotion prediction viewpoints according to the common experience group and the rich experience group are calculated according to the preset weights.
  • Calculate the investor's latest subjective sentiment predictions For example, analyze the general experience group and the rich experience group. After subjective emotions predict the point of view, the rich experience group can be given a higher weighting factor than the ordinary experience group, and the subjective sentiment prediction viewpoints of the common experience group and the rich experience group are combined to comprehensively determine the investor's latest subjective emotion prediction viewpoint.
  • the TOP50 vocabulary vector with the highest similarity among the TOP1000 vocabulary vectors of the general experience group and the rich experience group may be separately selected, and the subjective emotion prediction view of the keyword tag corresponding to each vocabulary vector is determined, and The rich experience group has a higher weighting factor than the ordinary experience group, and calculates the sum of the subjective sentiment predictions corresponding to all the vocabulary vectors of the TOP50 vocabulary vector of the general experience group and the TOP50 vocabulary vector of the rich experience group, and the judgment is ultimately the proportion of the long-term prediction viewpoint.
  • Large or short-term forecasting views account for a larger proportion, and the subjective sentiment forecasting view with a larger proportion is chosen as the investor's latest subjective sentiment forecasting perspective.
  • the investor is divided into a common experience group and a rich experience group, and the rich experience group is given a higher weight to determine the investor's latest subjective emotion prediction viewpoint, which can further improve the asset allocation income and reduce the risk.
  • FIG. 2 is a schematic flowchart of an embodiment of a social data-based asset configuration method according to an embodiment of the present application.
  • the social data-based asset configuration method includes the following steps:
  • Step S10 Obtain social data related to the preset category asset in the latest preset time period from the preset data source.
  • the social data related to the preset category asset in the latest preset time period of the investor is obtained from the preset data source, wherein the preset category asset is large.
  • assets as equity stocks, fixed investment bonds, alternative investment gold, derivatives option futures, etc.
  • the social data includes a comment or analysis text information of an investor in a preset range for a preset category asset within a preset preset time period, for example, a review of the investor's latest preset time period for different categories of assets, Analyze and other textual information, based on the LDA subject classification method, the investor's comments on different major assets (stocks, bonds, gold, options, futures, etc.) are classified and collected.
  • the social data of the investor's latest preset time period may also be obtained from the investor's QQ, Weibo, WeChat, Snowball, Oriental Fortune and other social software, including but not limited to Investors actively send out articles and circle of friends related to the pre-defined category assets, as well as comments from investors on the content posted by others, forwarding content, and so on.
  • the social data of the financial type can be increased when the social data of the latest preset time period of the investor is obtained, such as increasing the social software from the investor financial system (such as snowball, Financial social data obtained on Oriental Wealth, etc., and the amount or weight of financial social data related to financial issues issued by investors.
  • Step S20 Calculate and convert a number of vocabulary vectors based on the social data.
  • the preset points can be utilized. Word mode segmentation of the social data.
  • the word segmentation method of the string matching may be used to perform word segmentation processing on the social data, such as a forward maximum matching method, and the character string in one information is segmented from left to right, that is, the social data is left to right.
  • word segmentation processing such as a forward maximum matching method, and the character string in one information is segmented from left to right, that is, the social data is left to right.
  • Several consecutive characters in the table match the vocabulary, if the match, a word is segmented; or, the reverse maximum matching method, the character string in a message is segmented from right to left, that is, from the social data
  • the end of the matching scan starts, and several consecutive characters in the information text of the word to be distinguished are matched with the vocabulary from right to left.
  • Word lexical segmentation can also be used to classify each piece of information. Word lexical segmentation is a segmentation method for machine speech judgment. It uses syntactic information and semantic information to deal with ambiguity phenomena to segment words. The statistical word segmentation method can also be used for word segmentation processing. From the current user's historical search record or the public user's historical search record, according to the statistics of the phrase, some two adjacent words appear to have more frequent frequencies. These two adjacent words can be used as a phrase to perform word segmentation. In addition, word segmentation can also be performed based on the Chinese word segmentation tool NLPIR, which is not limited herein.
  • the word vector model word2vec is used to convert each word segmentation into a vocabulary vector, wherein word2vec is a tool for converting words into a vector form, which can simplify the processing of the text content into a vector space.
  • the vector operation calculates the similarity in the vector space to represent the semantic similarity of the text.
  • word2vec can be used to simplify the processing of text content into vector operations in K-dimensional vector space, and the similarity in vector space can be used to represent the semantic similarity of text. Therefore, the word vector output by word2vec can be used to do things like clustering, finding synonyms, part of speech analysis, etc. Moreover, word2vec is very efficient.
  • Step S30 Perform similarity calculation on each vocabulary vector and the preset keyword respectively, and determine that the keyword with the highest similarity with each vocabulary vector is the corresponding keyword of the vocabulary vector, and the keywords are pre-marked differently. Subjective emotions predict the point of view.
  • a topic dictionary that characterizes different subjective emotion prediction opinions (long, short, etc.) is constructed, and keywords in the dictionary are marked according to different subjective emotion prediction viewpoints.
  • the different subjective sentiment prediction views of the mark include long-term, short-term forecasting opinions and the like.
  • keywords such as “promising”, “strong momentum” and “seeing more” in the dictionary can be marked as “long” predictions; the key words in the dictionary are “not optimistic”, “lower”, “bearish” and so on. Words can be marked as "short” predictions.
  • the converted lexical vector may be similarly calculated with the keywords in the dictionary marked with different subjective emotion prediction views, for example, Word2vec converts the keywords in the dictionary marked with different subjective emotion prediction into vector form, so that the similarity calculation of several lexical vectors and keywords can be simplified to the vector operation in vector space, and the similarity in vector space can be calculated.
  • Degree which can be used to represent several vocabulary vectors and keywords. The semantic similarity of text.
  • the similarity calculation may be performed separately for each of the keywords in the keywords whose vocabulary vectors have different subjective emotion prediction views. For example, if the subjective sentiment forecasting perspective in the subjective sentiment predictive perspective dictionary includes “long” and “short” predictive opinions, the keywords that mark the "long” predictive perspective include “optimistic", “see more”, and mark “short” prediction views. Keywords include “lower” and “bearish”. After the user's social data is segmented and converted into several vocabulary vectors, each vocabulary vector of the converted vocabulary vectors can be similarly calculated with the keywords to determine the keywords with the highest similarity with each vocabulary vector. Is the corresponding keyword of the vocabulary vector.
  • the vocabulary vector A can be similarly calculated for each of the keywords of "optimistic", “too much”, “lower”, and “bearish”.
  • the similarity between the vocabulary vector A and each keyword can be calculated, and the keyword with the highest similarity with the vocabulary vector A is selected as the corresponding keyword of the vocabulary vector A, such as the vocabulary vector A and "optimistic", "see more", "
  • the similarities between "low” and “bearish” are 90%, 80%, 40%, and 30%, respectively. If the keyword with the highest similarity to vocabulary vector A is "optimistic”, then "optimistic" is used as the correspondence of vocabulary vector A. Key words.
  • step S40 all vocabulary vectors are sorted according to the similarity from high to low according to the corresponding keyword similarity, and a preset number of vocabulary vectors ranked first is selected.
  • the similarity between the vocabulary vector a and the corresponding keyword "good” is 99%
  • the similarity between the vocabulary vector b and the corresponding keyword "see more” is 98%
  • the vocabulary vector c is similar to the corresponding keyword "lower”.
  • the degree is 97%
  • the similarity between the vocabulary vector d and the corresponding keyword "bearish” is 96%
  • the order of similarity from high to low is a, b, c, d.
  • a preset number for example, 50
  • Step S50 Acquire a corresponding keyword of the preset number of vocabulary vectors, and determine a user's latest subjective sentiment prediction viewpoint according to the subjective sentiment prediction viewpoint marked by the corresponding keyword.
  • the vocabulary vector with the highest similarity to the keywords in the dictionary such as TOP50
  • the subjective sentiment prediction viewpoint marked by the corresponding keyword of the TOP50 vocabulary vector is selected to determine the investor's latest subjective emotion prediction viewpoint. For example, the number or proportion of long-term, short-term prediction opinions in the subjective emotion prediction viewpoint marked by the corresponding keyword of the selected TOP50 vocabulary vector can be counted, and the latest subjective emotion as the investor is selected in the selected quantity or the proportion Forecasting perspectives.
  • the number of long-predicted opinions in the subjective sentiment prediction point marked by the selected TOP50 vocabulary vector is 30, and the number of short-predicted opinions is 20, it means that the overall investor has a long-term forecasting viewpoint. For the majority, it is determined that the investor's latest subjective sentiment predictions are long-term predictions.
  • Step S60 adjusting the ratio of the preset category assets in the preset asset allocation based on the investor's latest subjective emotion prediction viewpoint.
  • the present embodiment adjusts preset asset types in the preset asset allocation based on the investor's latest subjective emotion prediction viewpoint.
  • preset asset types such as the proportion of large categories of assets. For example, if the investor's latest subjective sentiment forecasting view is a long-term forecasting viewpoint, increase the proportion of the pre-set asset allocation in the preset asset allocation; if the investor's latest subjective sentiment forecasting viewpoint is a short-term forecasting viewpoint, the pre-reduction is reduced.
  • real-time dynamic social data is introduced, and the investor's subjective tendency prediction for different assets is more accurate and real-time.
  • objective financial analysis basic analysis, technology
  • Indicator analysis while incorporating the overall investor's subjective emotional views and forecasting views, effectively improve the timeliness of asset allocation.
  • the present embodiment determines the emotional analysis and subjective expectation of the overall investor on the financial investment product by analyzing the social data of the latest preset time period of the investor, and based on the above-mentioned subjective influence factor of the investor. Perform asset allocation.
  • the step S50 includes:
  • Step S51 the investor corresponding to the social data is divided into a common experience group and a rich experience group according to a preset classification rule; for example, based on the historical investment data of the investor, the investment will be Investors with high return on investment and/or investment experience are divided into rich experience groups, and investors with low investment yield and/or low investment experience are divided into ordinary experience groups.
  • the similarity calculation is performed on a plurality of vocabulary vectors and keywords whose preset subjective emotion prediction views are different, and the vocabulary vectors with the highest similarity among the vocabulary vectors corresponding to the common experience group and the rich experience group are respectively selected. For example, the similarity calculations are compared between the different grouped vocabulary vectors and the keywords marked with different subjective emotion prediction views in the dictionary, and the vocabulary list with the highest similarity of the different experience groups TOP1000 is selected.
  • Step S52 analyzing subjective emotion prediction views of the common experience group and the rich experience group according to a preset number of vocabulary vectors according to a preset analysis rule, and predicting opinions according to subjective emotions of the common experience group and the rich experience group according to a preset
  • the weight calculation method calculates the investor's latest subjective emotion prediction view. For example, after analyzing the subjective sentiment predictions of the general experience group and the rich experience group, the rich experience group can be given a higher weighting factor than the ordinary experience group, and the subjective sentiment prediction views of the common experience group and the rich experience group can be integrated to comprehensively determine the investment. The latest subjective mood predictions.
  • the TOP50 vocabulary vector with the highest similarity among the TOP1000 vocabulary vectors of the general experience group and the rich experience group may be separately selected, and the subjective emotion prediction view of the keyword tag corresponding to each vocabulary vector is determined, and The rich experience group has a higher weighting factor than the ordinary experience group, and calculates the sum of the subjective sentiment predictions corresponding to all the vocabulary vectors of the TOP50 vocabulary vector of the general experience group and the TOP50 vocabulary vector of the rich experience group, and the judgment is ultimately the proportion of the long-term prediction viewpoint.
  • Large or short-term forecasting views account for a larger proportion, and the subjective sentiment forecasting view with a larger proportion is chosen as the investor's latest subjective sentiment forecasting perspective.
  • the investor is divided into a common experience group and a rich experience group, and the rich experience group is given a higher weight to determine the investor's latest subjective emotion prediction viewpoint, which can further improve the asset allocation income and reduce the risk.
  • the present application also provides a computer readable storage medium storing a social data based asset configuration system, the social data based asset configuration system executable by at least one processor to enable The at least one processor performs the steps of the social data-based asset configuration method as in the above embodiment, and the specific implementation processes of the steps S10, S20, S30, etc. of the social data-based asset configuration method are as described above, and are not Let me repeat.
  • the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and can also be implemented by hardware, but in many cases, the former is A better implementation.
  • the technical solution of the present application which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk,
  • the optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Artificial Intelligence (AREA)
  • Accounting & Taxation (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Human Resources & Organizations (AREA)
  • Operations Research (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • General Business, Economics & Management (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种基于社交数据的资产配置方法、电子装置（1）及介质，该方法包括：从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据（S10）；基于社交数据计算转换得到若干词汇向量（S20）；将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为词汇向量的对应关键词（S30）；根据对应关键词相似度将所有词汇向量按相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量（S40）；获取预设数量的词汇向量的对应关键词，并根据对应关键词所标记的主观情绪预测观点确定最新主观情绪预测观点（S50）；基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况（S60）。该方法能提升资产配置的时效性和全面性。

Description

基于社交数据的资产配置方法、电子装置及介质
本申请基于巴黎公约申明享有2017年9月5日递交的申请号为CN 201710790355.9、名称为“基于社交数据的资产配置方法、电子装置及介质”中国专利申请的优先权，该中国专利申请的整体内容以参考的方式结合在本申请中。
技术领域
本申请涉及计算机技术领域，尤其涉及一种基于社交数据的资产配置方法、电子装置及介质。
背景技术
现有大数据资产配置模型主要基于不同资产的基本面分析、技术指标分析来设计有效因子，构建量化资产模型，预测不同类型资产的风险与收益，并最终制定合适投资者的资产组合。然而，现有的资产配置模型只能单纯基于客观数据(如基本面分析、技术指标分析)来进行资产配置，没有考虑到投资者的主观影响因子，在时效性和全面性上存在缺失。
发明内容
本申请的目的在于提供一种基于社交数据的资产配置方法、电子装置及介质，旨在提高资产配置的时效性。
为实现上述目的，本申请第一方面提供一种电子装置，所述电子装置包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的基于社交数据的资产配置系统，所述基于社交数据的资产配置系统被所述处理器执行时实现如下步骤：
A、从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据；
B、基于所述社交数据计算转换得到若干词汇向量；
C、将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词，所述关键词预先标记有不同主观情绪预测观点；
D、根据对应关键词相似度将所有词汇向量按所述相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量；
E、获取所述预设数量的词汇向量的对应关键词，并根据所述对应关键词所标记的主观情绪预测观点确定投资者的最新主观情绪预测观点；
F、基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况。
本申请第二方面提供一种基于社交数据的资产配置方法，所述基于社交数据的资产配置方法包括：
A、从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据；
B、基于所述社交数据计算转换得到若干词汇向量；
C、将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词，所述关键词预先标记有不同主观情绪预测观点；
D、根据对应关键词相似度将所有词汇向量按所述相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量；
E、获取所述预设数量的词汇向量的对应关键词，并根据所述对应关键词所标记的主观情绪预测观点确定投资者的最新主观情绪预测观点；
F、基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况。
本申请第三方面提供一种计算机可读存储介质，所述计算机可读存储介质存储有基于社交数据的资产配置系统，所述基于社交数据的资产配置系统可被至少一个处理器执行，以使所述至少一个处理器执行如上述的基于社交数据的资产配置方法的步骤。
本申请提出的基于社交数据的资产配置方法、电子装置及介质，通过分析挖掘投资者最新预设时间段的社交数据，确定整体投资者在金融投资产品上的情感分析和主观预期，并基于投资者的上述主观影响因子来进行资产配置。由于引入实时动态的社交数据，来进行整体投资者在金融投资产品上的情感分析和主观预期；并在最终资产配置方案上，在考虑不同资产配置过程中融合整体投资者在社交数据反馈的情绪和主观预期，从而提升资产配置的时效性和全面性。
附图说明
图1为本申请资产配置系统10较佳实施例的运行环境示意图；
图2为本申请基于社交数据的资产配置方法一实施例的流程示意图；
图3为图2所示步骤S50的细化流程示意图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合 附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本申请，并不用于限定本申请。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
需要说明的是，在本申请中涉及“第一”、“第二”等的描述仅用于描述目的，而不能理解为指示或暗示其相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。另外，各个实施例之间的技术方案可以相互结合，但是必须是以本领域普通技术人员能够实现为基础，当技术方案的结合出现相互矛盾或无法实现时应当认为这种技术方案的结合不存在，也不在本申请要求的保护范围之内。
本申请提供一种基于社交数据的资产配置系统。请参阅图1，是本申请资产配置系统10较佳实施例的运行环境示意图。
在本实施例中，所述的资产配置系统10安装并运行于电子装置1中。该电子装置1可包括，但不仅限于，存储器11、处理器12及显示器13。图1仅示出了具有组件11-13的电子装置1，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。
所述存储器11至少包括一种类型的可读存储介质，所述存储器11在一些实施例中可以是所述电子装置1的内部存储单元，例如该电子装置1的硬盘或内存。所述存储器11在另一些实施例中也可以是所述电子装置1的外部存储设备，例如所述电子装置1上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。进一步地，所述存储器11还可以既包括所述电子装置1的内部存储单元也包括外部存储设备。所述存储器11用于存储安装于所述电子装置1的应用软件及各类数据，例如所述资产配置系统10的程序代码等。所述存储器11还可以用于暂时地存储已经输出或者将要输出的数据。
所述处理器12在一些实施例中可以是一中央处理器(Central Processing Unit,CPU)，微处理器或其他数据处理芯片，用于运行所述存储器11中存储的程序代码或处理数据，例如执行所述资产配置系统10等。
所述显示器13在一些实施例中可以是LED显示器、液晶显示器、触控式液晶显示器以及OLED(Organic Light-Emitting Diode，有机发光二极管)触摸器等。所述显示器13用于显示在所述电子装置1中处理的信息以及用于显示可视化的用户界面，例如确定出的投资者最新主观情绪预测观点、资产配置中不同类别资产的配比情况等。所述电子装置1的部件11-13通过系统总线相互通信。
基于社交数据的资产配置系统10包括至少一个存储在所述存储器11中的计算机可读指令，该至少一个计算机可读指令可被所述处理器12执行，以实现本申请各实施例。
其中，上述基于社交数据的资产配置系统10被所述处理器12执行时实现如下步骤：
步骤S1，从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据。
本实施例中，在接收到资产配置请求时，首先从预设的数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据，其中，所述预设类别资产为大类资产如权益类股票、固定投资类债券、另类投资黄金、衍生品期权期货等。所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息，如可整理投资者最新预设时间段内针对不同大类资产的评论、分析等文本信息，基于LDA主题分类方法将投资者针对不同大类资产(股票、债券、黄金、期权、期货等)的评论进行主题分类收集。还可从投资者的QQ、微博、微信、雪球、东方财富等社交软件上获取投资者最新预设时间段如最近3个月、6个月的社交数据，该社交数据包括但不限于投资者主动发出的与预设类别资产相关的文章、朋友圈等内容，以及投资者对其他人发布内容的评论、转发内容，等等。
进一步地，由于涉及金融产品的资产配置，因此，在获取投资者最新预设时间段的社交数据时，可增加金融类社交数据的权重，如增加从投资者金融类社交软件(如雪球、东方财富等)上获取的金融社交数据以及投资者发出的与金融类相关的金融社交数据的数量或权重。
步骤S2，基于所述社交数据计算转换得到若干词汇向量。
在获取到投资者最新预设时间段的社交数据后，可利用预设的分词方式对所述社交数据进行分词。例如，可利用字符串匹配的分词方法对所述社交数据进行分词处理，如正向最大匹配法，把一个信息中的字符串从左至右来分词，即从左到右将所述社交数据中的几个连续字符与词表匹配，如果匹配上，则切分出一个词；或者，反向最大匹配法，把一个信息中的字符串从右至左来分词，即从所述社交数据的末端开始匹配扫描，从右至左将待分词的信息文本中的几个连续字符与词表匹配，如果匹配上，则切分出一个词；或者，最短路径分词法，一个信息中的字符串里面要求切出的词数是最少的；或者，双向最大匹配法，正反向同时进行分词匹配。还可利用词义分词法对各个信息进行分词处理，词义分词法是一种机器语音判断的分词方法，利用句法信息和语义信息来处理歧义现象来分词。还可利用统计分词法对各个信息进行分词处理，从当前用户的历史搜索记录或大众用户的历史 搜索记录中，根据词组的统计，会统计有些两个相邻的字出现的频率较多，则可将这两个相邻的字作为词组来进行分词。此外，还可基于中文分词工具NLPIR进行分词，在此不做限定。
对所述社交数据进行分词后，采用词向量模型word2vec将各个分词计算转换为词汇向量，其中，word2vec是一个将单词转换成向量形式的工具，可以把对文本内容的处理简化为向量空间中的向量运算，计算出向量空间上的相似度，来表示文本语义上的相似度。例如，word2vec通过训练，可以把对文本内容的处理简化为K维向量空间中的向量运算，而向量空间上的相似度可以用来表示文本语义上的相似度。因此，word2vec输出的词向量可以被用来做如聚类、找同义词、词性分析等等工作，而且，word2vec非常高效。
步骤S3，将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词，所述关键词预先标记有不同主观情绪预测观点。
本实施例中，首先构建表征客户不同主观情绪预测观点(多头、空头等)的主题词典，并依据不同主观情绪预测观点对词典中的关键词进行标记。其中，标记的不同主观情绪预测观点包括多头、空头预测观点等等。例如，对词典中的“看好”、“势头强劲”、“看多”等关键词，可标记为“多头”预测观点；词典中的“不看好”、“看低”、“看跌”等关键词，可标记为“空头”预测观点。
在对整体投资者的社交数据进行分词，并将各个分词计算转换为词汇向量后，可将转换的若干词汇向量与词典中标记有不同主观情绪预测观点的关键词进行相似度计算，例如，可通过word2vec将词典中标记有不同主观情绪预测观点的关键词转换成向量形式，这样，可将若干词汇向量与关键词的相似度计算简化为向量空间中的向量运算，计算出向量空间上的相似度，即可用来表示若干词汇向量与关键词在文本语义上的相似度。
具体地，可将若干词汇向量与预设的标记有不同主观情绪预测观点的关键词中每一关键词分别进行相似度计算。例如，若主观情绪预测观点词典中的主观情绪预测观点包括“多头”和“空头”预测观点，标记“多头”预测观点的关键词包括“看好”、“看多”，标记“空头”预测观点的关键词包括“走低”、“看跌”。对用户的社交数据进行分词、转换成若干词汇向量后，可将转换得到的若干词汇向量中每一词汇向量与关键词分别进行相似度计算，以确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词。例如，针对词汇向量A，可将词汇向量A与“看好”、“看多”、“走低”、“看跌”中的每一个关键词分别进行相似度计算。可计算出词汇向量A与每个关键词的相似度，选择与词汇向量A相似度最高的关键词作为词汇向 量A的对应关键词，如词汇向量A与“看好”、“看多”、“走低”、“看跌”的相似度分别为90％、80％、40％、30％，则与词汇向量A相似度最高的关键词为“看好”，则将“看好”作为词汇向量A的对应关键词。
步骤S4，根据对应关键词相似度将所有词汇向量按所述相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量。
例如，若词汇向量a与对应关键词“看好”的相似度为99％，词汇向量b与对应关键词“看多”的相似度为98％，词汇向量c与对应关键词“走低”的相似度为97％，词汇向量d与对应关键词“看跌”的相似度为96％，则按相似度从高到低排序依次为a、b、c、d。以此，可根据相似度计算结果挑选出排序靠前即与关键词的相似度最高的预设数量(如50个)词汇向量。
步骤S5，获取所述预设数量的词汇向量的对应关键词，并根据所述对应关键词所标记的主观情绪预测观点确定用户的最新主观情绪预测观点。
获取所述预设数量(如50个)的词汇向量的对应关键词，并可根据预设数量(如50个)词汇向量的对应关键词所标记的主观情绪预测观点确定该用户的最新主观情绪预测观点。例如，挑选出的50个词汇向量的对应关键词中有10个“看好”、10个“看多”、20个“走低”和10个“看跌”，则可根据关键词标记的主观情绪预测观点确定50个词汇向量中包含“多头”主观情绪预测观点的词汇有20个，包含“空头”主观情绪预测观点的词汇有30个，“空头”主观情绪预测观点的词汇数量比“多头”主观情绪预测观点的词汇数量多，则可以此确定该用户的最新主观情绪预测观点为“空头”主观情绪预测观点。
计算出若干词汇向量与词典中关键词的相似度之后，可挑选出与词典中关键词的相似度最高的预设数量如TOP50的词汇向量，并根据挑选出的TOP50的词汇向量的对应关键词所标记的主观情绪预测观点来确定投资者的最新主观情绪预测观点。例如，可统计挑选出的TOP50的词汇向量的对应关键词所标记的主观情绪预测观点中多头、空头预测观点的数量或所占比例，选择数量或所占比例高的作为投资者的最新主观情绪预测观点。例如，若挑选出的TOP50的词汇向量对应的关键词所标记的主观情绪预测观点中多头预测观点的数量为30个，空头预测观点的数量为20个，则说明整体投资者中持多头预测观点的占大多数，则确定投资者的最新主观情绪预测观点为多头预测观点。
步骤S6，基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况。
由于考虑到金融市场是相对动态的市场，基本由多空双方力量博 弈决定。投资者针对不同大类资产主观的预期判断以及情感分析，同样会影响不同资产类型的风险和收益。因此，本实施例在基于客观数据(如基本面分析、技术指标分析)来进行预设的资产配置基础上，还基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产如大类资产的配比情况。例如，若投资者的最新主观情绪预测观点为多头预测观点，则增加预设的资产配置中预设类别资产的配置占比；若投资者的最新主观情绪预测观点为空头预测观点，则减少预设的资产配置中预设类别资产的配置占比。
本实施例中，引入实时动态的社交数据，对于投资者对于不同资产的主观倾向预测更加精准和实时，在大类资产配置调整过程中，不仅仅考虑客观的金融数据分析(基本面分析、技术指标分析)，同时融入整体投资者在主观上的情绪观点和预测观点，有效地提升资产配置的时效性。
与现有技术相比，本实施例通过分析挖掘投资者最新预设时间段的社交数据，确定整体投资者在金融投资产品上的情感分析和主观预期，并基于投资者的上述主观影响因子来进行资产配置。由于引入实时动态的社交数据，来进行整体投资者在金融投资产品上的情感分析和主观预期；并在最终资产配置方案上，在考虑不同资产配置过程中融合整体投资者在社交数据反馈的情绪和主观预期，从而提升资产配置的时效性和全面性。
在一可选的实施例中，在上述图1的实施例的基础上，所述基于社交数据的资产配置系统10被所述处理器12执行实现所述步骤S5时，具体包括：
按预设分类规则将所述社交数据对应的投资者区分为普通经验组和丰富经验组；如，可基于投资者的历史投资数据，将投资收益率高和/或投资经验丰富的投资者划分至丰富经验组，将投资收益率低和/或投资经验少的投资者划分至普通经验组。
将若干词汇向量与预设的标记有不同主观情绪预测观点的关键词进行相似度计算，分别挑选出普通经验组、丰富经验组对应的词汇向量中相似度最高的预设数量的词汇向量。如将不同分组词汇向量和词典中标记有不同主观情绪预测观点的关键词进行相似度计算比对，挑选出不同经验组TOP1000相似度最大的词汇列表。
根据预设数量的词汇向量按预设的分析规则分别分析出普通经验组、丰富经验组的主观情绪预测观点，并根据普通经验组、丰富经验组的主观情绪预测观点按预设的权重计算方式计算得到投资者的最新主观情绪预测观点。例如，分析出普通经验组、丰富经验组的 主观情绪预测观点后，可赋予丰富经验组相对普通经验组更高的权重因子，融合普通经验组和丰富经验组的主观情绪预测观点来综合确定投资者的最新主观情绪预测观点。
在另一种实施方式中，还可分别挑选出普通经验组、丰富经验组的TOP1000词汇向量中相似度最高的TOP50词汇向量，确定每一词汇向量对应的关键词标记的主观情绪预测观点，赋予丰富经验组相对普通经验组更高的权重因子，计算普通经验组的TOP50词汇向量与丰富经验组的TOP50词汇向量的所有词汇对应的主观情绪预测观点总和，判断最终是多头预测观点所占比例更大，还是空头预测观点所占比例更大，选择所占比例更大的主观情绪预测观点作为投资者的最新主观情绪预测观点。
本实施例中将投资者区分为普通经验组和丰富经验组，并赋予丰富经验组更高的权重来确定投资者的最新主观情绪预测观点，能进一步提高资产配置的收益及降低风险。
如图2所示，图2为本申请基于社交数据的资产配置方法一实施例的流程示意图，该基于社交数据的资产配置方法包括以下步骤：
步骤S10，从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据。
本实施例中，在接收到资产配置请求时，首先从预设的数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据，其中，所述预设类别资产为大类资产如权益类股票、固定投资类债券、另类投资黄金、衍生品期权期货等。所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息，如可整理投资者最新预设时间段内针对不同大类资产的评论、分析等文本信息，基于LDA主题分类方法将投资者针对不同大类资产(股票、债券、黄金、期权、期货等)的评论进行主题分类收集。还可从投资者的QQ、微博、微信、雪球、东方财富等社交软件上获取投资者最新预设时间段如最近3个月、6个月的社交数据，该社交数据包括但不限于投资者主动发出的与预设类别资产相关的文章、朋友圈等内容，以及投资者对其他人发布内容的评论、转发内容，等等。
进一步地，由于涉及金融产品的资产配置，因此，在获取投资者最新预设时间段的社交数据时，可增加金融类社交数据的权重，如增加从投资者金融类社交软件(如雪球、东方财富等)上获取的金融社交数据以及投资者发出的与金融类相关的金融社交数据的数量或权重。
步骤S20，基于所述社交数据计算转换得到若干词汇向量。
在获取到投资者最新预设时间段的社交数据后，可利用预设的分 词方式对所述社交数据进行分词。例如，可利用字符串匹配的分词方法对所述社交数据进行分词处理，如正向最大匹配法，把一个信息中的字符串从左至右来分词，即从左到右将所述社交数据中的几个连续字符与词表匹配，如果匹配上，则切分出一个词；或者，反向最大匹配法，把一个信息中的字符串从右至左来分词，即从所述社交数据的末端开始匹配扫描，从右至左将待分词的信息文本中的几个连续字符与词表匹配，如果匹配上，则切分出一个词；或者，最短路径分词法，一个信息中的字符串里面要求切出的词数是最少的；或者，双向最大匹配法，正反向同时进行分词匹配。还可利用词义分词法对各个信息进行分词处理，词义分词法是一种机器语音判断的分词方法，利用句法信息和语义信息来处理歧义现象来分词。还可利用统计分词法对各个信息进行分词处理，从当前用户的历史搜索记录或大众用户的历史搜索记录中，根据词组的统计，会统计有些两个相邻的字出现的频率较多，则可将这两个相邻的字作为词组来进行分词。此外，还可基于中文分词工具NLPIR进行分词，在此不做限定。
对所述社交数据进行分词后，采用词向量模型word2vec将各个分词计算转换为词汇向量，其中，word2vec是一个将单词转换成向量形式的工具，可以把对文本内容的处理简化为向量空间中的向量运算，计算出向量空间上的相似度，来表示文本语义上的相似度。例如，word2vec通过训练，可以把对文本内容的处理简化为K维向量空间中的向量运算，而向量空间上的相似度可以用来表示文本语义上的相似度。因此，word2vec输出的词向量可以被用来做如聚类、找同义词、词性分析等等工作，而且，word2vec非常高效。
步骤S30，将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词，所述关键词预先标记有不同主观情绪预测观点。
本实施例中，首先构建表征客户不同主观情绪预测观点(多头、空头等)的主题词典，并依据不同主观情绪预测观点对词典中的关键词进行标记。其中，标记的不同主观情绪预测观点包括多头、空头预测观点等等。例如，对词典中的“看好”、“势头强劲”、“看多”等关键词，可标记为“多头”预测观点；词典中的“不看好”、“看低”、“看跌”等关键词，可标记为“空头”预测观点。
在对整体投资者的社交数据进行分词，并将各个分词计算转换为词汇向量后，可将转换的若干词汇向量与词典中标记有不同主观情绪预测观点的关键词进行相似度计算，例如，可通过word2vec将词典中标记有不同主观情绪预测观点的关键词转换成向量形式，这样，可将若干词汇向量与关键词的相似度计算简化为向量空间中的向量运算，计算出向量空间上的相似度，即可用来表示若干词汇向量与关键词在 文本语义上的相似度。
具体地，可将若干词汇向量与预设的标记有不同主观情绪预测观点的关键词中每一关键词分别进行相似度计算。例如，若主观情绪预测观点词典中的主观情绪预测观点包括“多头”和“空头”预测观点，标记“多头”预测观点的关键词包括“看好”、“看多”，标记“空头”预测观点的关键词包括“走低”、“看跌”。对用户的社交数据进行分词、转换成若干词汇向量后，可将转换得到的若干词汇向量中每一词汇向量与关键词分别进行相似度计算，以确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词。例如，针对词汇向量A，可将词汇向量A与“看好”、“看多”、“走低”、“看跌”中的每一个关键词分别进行相似度计算。可计算出词汇向量A与每个关键词的相似度，选择与词汇向量A相似度最高的关键词作为词汇向量A的对应关键词，如词汇向量A与“看好”、“看多”、“走低”、“看跌”的相似度分别为90％、80％、40％、30％，则与词汇向量A相似度最高的关键词为“看好”，则将“看好”作为词汇向量A的对应关键词。
步骤S40，根据对应关键词相似度将所有词汇向量按所述相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量。
例如，若词汇向量a与对应关键词“看好”的相似度为99％，词汇向量b与对应关键词“看多”的相似度为98％，词汇向量c与对应关键词“走低”的相似度为97％，词汇向量d与对应关键词“看跌”的相似度为96％，则按相似度从高到低排序依次为a、b、c、d。以此，可根据相似度计算结果挑选出排序靠前即与关键词的相似度最高的预设数量(如50个)词汇向量。
步骤S50，获取所述预设数量的词汇向量的对应关键词，并根据所述对应关键词所标记的主观情绪预测观点确定用户的最新主观情绪预测观点。
获取所述预设数量(如50个)的词汇向量的对应关键词，并可根据预设数量(如50个)词汇向量的对应关键词所标记的主观情绪预测观点确定该用户的最新主观情绪预测观点。例如，挑选出的50个词汇向量的对应关键词中有10个“看好”、10个“看多”、20个“走低”和10个“看跌”，则可根据关键词标记的主观情绪预测观点确定50个词汇向量中包含“多头”主观情绪预测观点的词汇有20个，包含“空头”主观情绪预测观点的词汇有30个，“空头”主观情绪预测观点的词汇数量比“多头”主观情绪预测观点的词汇数量多，则可以此确定该用户的最新主观情绪预测观点为“空头”主观情绪预测观点。
计算出若干词汇向量与词典中关键词的相似度之后，可挑选出与词典中关键词的相似度最高的预设数量如TOP50的词汇向量，并根据 挑选出的TOP50的词汇向量的对应关键词所标记的主观情绪预测观点来确定投资者的最新主观情绪预测观点。例如，可统计挑选出的TOP50的词汇向量的对应关键词所标记的主观情绪预测观点中多头、空头预测观点的数量或所占比例，选择数量或所占比例高的作为投资者的最新主观情绪预测观点。例如，若挑选出的TOP50的词汇向量对应的关键词所标记的主观情绪预测观点中多头预测观点的数量为30个，空头预测观点的数量为20个，则说明整体投资者中持多头预测观点的占大多数，则确定投资者的最新主观情绪预测观点为多头预测观点。
步骤S60，基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况。
由于考虑到金融市场是相对动态的市场，基本由多空双方力量博弈决定。投资者针对不同大类资产主观的预期判断以及情感分析，同样会影响不同资产类型的风险和收益。因此，本实施例在基于客观数据(如基本面分析、技术指标分析)来进行预设的资产配置基础上，还基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产如大类资产的配比情况。例如，若投资者的最新主观情绪预测观点为多头预测观点，则增加预设的资产配置中预设类别资产的配置占比；若投资者的最新主观情绪预测观点为空头预测观点，则减少预设的资产配置中预设类别资产的配置占比。
本实施例中，引入实时动态的社交数据，对于投资者对于不同资产的主观倾向预测更加精准和实时，在大类资产配置调整过程中，不仅仅考虑客观的金融数据分析(基本面分析、技术指标分析)，同时融入整体投资者在主观上的情绪观点和预测观点，有效地提升资产配置的时效性。
与现有技术相比，本实施例通过分析挖掘投资者最新预设时间段的社交数据，确定整体投资者在金融投资产品上的情感分析和主观预期，并基于投资者的上述主观影响因子来进行资产配置。由于引入实时动态的社交数据，来进行整体投资者在金融投资产品上的情感分析和主观预期；并在最终资产配置方案上，在考虑不同资产配置过程中融合整体投资者在社交数据反馈的情绪和主观预期，从而提升资产配置的时效性和全面性。
在一可选的实施例中，如图3所示，在上述图2的实施例的基础上，所述步骤S50包括：
步骤S51，按预设分类规则将所述社交数据对应的投资者区分为普通经验组和丰富经验组；如，可基于投资者的历史投资数据，将投 资收益率高和/或投资经验丰富的投资者划分至丰富经验组，将投资收益率低和/或投资经验少的投资者划分至普通经验组。
将若干词汇向量与预设的标记有不同主观情绪预测观点的关键词进行相似度计算，分别挑选出普通经验组、丰富经验组对应的词汇向量中相似度最高的预设数量的词汇向量。如将不同分组词汇向量和词典中标记有不同主观情绪预测观点的关键词进行相似度计算比对，挑选出不同经验组TOP1000相似度最大的词汇列表。
步骤S52，根据预设数量的词汇向量按预设的分析规则分别分析出普通经验组、丰富经验组的主观情绪预测观点，并根据普通经验组、丰富经验组的主观情绪预测观点按预设的权重计算方式计算得到投资者的最新主观情绪预测观点。例如，分析出普通经验组、丰富经验组的主观情绪预测观点后，可赋予丰富经验组相对普通经验组更高的权重因子，融合普通经验组和丰富经验组的主观情绪预测观点来综合确定投资者的最新主观情绪预测观点。
在另一种实施方式中，还可分别挑选出普通经验组、丰富经验组的TOP1000词汇向量中相似度最高的TOP50词汇向量，确定每一词汇向量对应的关键词标记的主观情绪预测观点，赋予丰富经验组相对普通经验组更高的权重因子，计算普通经验组的TOP50词汇向量与丰富经验组的TOP50词汇向量的所有词汇对应的主观情绪预测观点总和，判断最终是多头预测观点所占比例更大，还是空头预测观点所占比例更大，选择所占比例更大的主观情绪预测观点作为投资者的最新主观情绪预测观点。
本实施例中将投资者区分为普通经验组和丰富经验组，并赋予丰富经验组更高的权重来确定投资者的最新主观情绪预测观点，能进一步提高资产配置的收益及降低风险。
此外，本申请还提供一种计算机可读存储介质，所述计算机可读存储介质存储有基于社交数据的资产配置系统，所述基于社交数据的资产配置系统可被至少一个处理器执行，以使所述至少一个处理器执行如上述实施例中的基于社交数据的资产配置方法的步骤，该基于社交数据的资产配置方法的步骤S10、S20、S30等具体实施过程如上文所述，在此不再赘述。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定 的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件来实现，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
以上参照附图说明了本申请的优选实施例，并非因此局限本申请的权利范围。上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。另外，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
本领域技术人员不脱离本申请的范围和实质，可以有多种变型方案实现本申请，比如作为一个实施例的特征可用于另一实施例而得到又一实施例。凡在运用本申请的技术构思之内所作的任何修改、等同替换和改进，均应在本申请的权利范围之内。

Claims (20)

  1. 一种电子装置，其特征在于，所述电子装置包括存储器、处理器，所述存储器上存储有可在所述处理器上运行的基于社交数据的资产配置系统，所述基于社交数据的资产配置系统被所述处理器执行时实现如下步骤：
    A1、从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据；
    B1、基于所述社交数据计算转换得到若干词汇向量；
    C1、将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词，所述关键词预先标记有不同主观情绪预测观点；
    D1、根据对应关键词相似度将所有词汇向量按所述相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量；
    E1、获取所述预设数量的词汇向量的对应关键词，并根据所述对应关键词所标记的主观情绪预测观点确定投资者的最新主观情绪预测观点；
    F1、基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况。
  2. 如权利要求1所述的电子装置，其特征在于，所述基于社交数据的资产配置系统被所述处理器执行实现所述步骤E1时，包括：
    按预设分类规则将所述社交数据对应的投资者区分为普通经验组和丰富经验组；
    根据预设数量的词汇向量按预设的分析规则分别分析出普通经验组、丰富经验组的主观情绪预测观点，并根据普通经验组、丰富经验组的主观情绪预测观点按预设的权重计算方式计算得到投资者的最新主观情绪预测观点。
  3. 如权利要求1所述的电子装置，其特征在于，标记的不同主观情绪预测观点包括多头预测观点或空头预测观点，所述基于社交数据的资产配置系统被所述处理器执行实现所述步骤F1时，包括：
    若投资者的最新主观情绪预测观点为多头预测观点，则增加预设的资产配置中预设类别资产的配置占比；若投资者的最新主观情绪预测观点为空头预测观点，则减少预设的资产配置中预设类别资产的配置占比。
  4. 如权利要求1所述的电子装置，其特征在于，所述社交数据 包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  5. 如权利要求2所述的电子装置，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  6. 如权利要求3所述的电子装置，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  7. 如权利要求1所述的电子装置，其特征在于，所述步骤B1包括：
    采用词向量模型word2vec将所述社交数据计算转换为若干词汇向量。
  8. 一种基于社交数据的资产配置方法，其特征在于，所述基于社交数据的资产配置方法包括：
    A2、从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据；
    B2、基于所述社交数据计算转换得到若干词汇向量；
    C2、将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词，所述关键词预先标记有不同主观情绪预测观点；
    D2、根据对应关键词相似度将所有词汇向量按所述相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量；
    E2、获取所述预设数量的词汇向量的对应关键词，并根据所述对应关键词所标记的主观情绪预测观点确定投资者的最新主观情绪预测观点；
    F2、基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况。
  9. 如权利要求8所述的基于社交数据的资产配置方法，其特征在于，所述步骤E2包括：
    按预设分类规则将所述社交数据对应的投资者区分为普通经验组和丰富经验组；
    根据预设数量的词汇向量按预设的分析规则分别分析出普通经验组、丰富经验组的主观情绪预测观点，并根据普通经验组、丰富经验组的主观情绪预测观点按预设的权重计算方式计算得到投资者的最新主观情绪预测观点。
  10. 如权利要求8所述的基于社交数据的资产配置方法，其特征在于，标记的不同主观情绪预测观点包括多头预测观点或空头预测观点，所述步骤F2包括：
    若投资者的最新主观情绪预测观点为多头预测观点，则增加预设的资产配置中预设类别资产的配置占比；若投资者的最新主观情绪预测观点为空头预测观点，则减少预设的资产配置中预设类别资产的配置占比。
  11. 如权利要求8所述的基于社交数据的资产配置方法，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  12. 如权利要求9所述的基于社交数据的资产配置方法，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  13. 如权利要求10所述的基于社交数据的资产配置方法，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  14. 如权利要求8所述的基于社交数据的资产配置方法，其特征在于，所述步骤B2包括：
    采用词向量模型word2vec将所述社交数据计算转换为若干词汇向量。
  15. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有基于社交数据的资产配置系统，所述基于社交数据的资产配置系统可被至少一个处理器执行，以使所述至少一个处理器执行如下步骤：
    A3、从预设数据源获取投资者在最新预设时间段内与预设类别资产相关的社交数据；
    B3、基于所述社交数据计算转换得到若干词汇向量；
    C3、将每一词汇向量与预设的关键词分别进行相似度计算，确定与每一词汇向量相似度最高的关键词为所述词汇向量的对应关键词，所述关键词预先标记有不同主观情绪预测观点；
    D3、根据对应关键词相似度将所有词汇向量按所述相似度由高到低进行排序，挑选出排序靠前的预设数量的词汇向量；
    E3、获取所述预设数量的词汇向量的对应关键词，并根据所述对应关键词所标记的主观情绪预测观点确定投资者的最新主观情绪预测观点；
    F3、基于投资者的最新主观情绪预测观点调整预设的资产配置中预设类别资产的配比情况。
  16. 如权利要求15所述的计算机可读存储介质，其特征在于，所述基于社交数据的资产配置系统被所述处理器执行实现所述步骤E3时，包括：
    按预设分类规则将所述社交数据对应的投资者区分为普通经验组和丰富经验组；
    根据预设数量的词汇向量按预设的分析规则分别分析出普通经验组、丰富经验组的主观情绪预测观点，并根据普通经验组、丰富经验组的主观情绪预测观点按预设的权重计算方式计算得到投资者的最新主观情绪预测观点。
  17. 如权利要求15所述的计算机可读存储介质，其特征在于，标记的不同主观情绪预测观点包括多头预测观点或空头预测观点，所述基于社交数据的资产配置系统被所述处理器执行实现所述步骤F3时，包括：
    若投资者的最新主观情绪预测观点为多头预测观点，则增加预设的资产配置中预设类别资产的配置占比；若投资者的最新主观情绪预测观点为空头预测观点，则减少预设的资产配置中预设类别资产的配置占比。
  18. 如权利要求15所述的计算机可读存储介质，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  19. 如权利要求16所述的计算机可读存储介质，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预 设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
  20. 如权利要求17所述的计算机可读存储介质，其特征在于，所述社交数据包括预设范围内的投资者在最新预设时间段内针对预设类别资产的评论或分析文本信息；所述预设类别资产包括股票、债券、黄金和/或期权期货。
PCT/CN2017/108795 2017-09-05 2017-10-31 基于社交数据的资产配置方法、电子装置及介质 Ceased WO2019047352A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710790355.9 2017-09-05
CN201710790355.9A CN107767273B (zh) 2017-09-05 2017-09-05 基于社交数据的资产配置方法、电子装置及介质

Publications (1)

Publication Number Publication Date
WO2019047352A1 true WO2019047352A1 (zh) 2019-03-14

Family

ID=61265640

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/108795 Ceased WO2019047352A1 (zh) 2017-09-05 2017-10-31 基于社交数据的资产配置方法、电子装置及介质

Country Status (2)

Country Link
CN (1) CN107767273B (zh)
WO (1) WO2019047352A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108733782A (zh) * 2018-05-08 2018-11-02 平安科技（深圳）有限公司 资产走势分析的方法、装置、计算机设备和存储介质
CN109102398A (zh) * 2018-06-21 2018-12-28 平安科技（深圳）有限公司 债券etf的构建方法、装置及计算机可读存储介质
CN109213934A (zh) * 2018-08-23 2019-01-15 阿里巴巴集团控股有限公司 一种资源的处理方法、装置及设备
CN109460550A (zh) * 2018-10-22 2019-03-12 平安科技（深圳）有限公司 利用大数据的证券研报情感分析方法、装置及计算机设备
CN112862617A (zh) * 2019-11-27 2021-05-28 泰康保险集团股份有限公司 数据处理方法、系统、存储介质及电子设备
CN114240672B (zh) * 2021-12-15 2023-08-18 平安科技（深圳）有限公司 绿色资产的占比的识别方法及相关产品
CN119474194B (zh) * 2025-01-17 2025-04-08 卓望数码技术（深圳）有限公司 虚拟数据向量化管理方法、装置、电子设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2618651A1 (en) * 2008-02-25 2009-08-25 Gerard Voon Quantum computing h/w and s/w and artificial intelligence
CN102054116A (zh) * 2009-10-30 2011-05-11 财团法人资讯工业策进会 情绪分析方法、情绪分析系统及情绪分析装置
CN104951807A (zh) * 2015-07-10 2015-09-30 沃民高新科技（北京）股份有限公司 股市情绪的确定方法和装置
CN104995650A (zh) * 2011-12-27 2015-10-21 汤姆森路透社全球资源公司 用于使用源于社交媒体的数据和情绪分析来生成复合索引的方法及系统
CN106095777A (zh) * 2016-05-26 2016-11-09 优品财富管理有限公司 基于大数据的预测证券市场多空情绪指标方法
US20170018033A1 (en) * 2015-07-15 2017-01-19 Foundation Of Soongsil University Industry Cooperation Stock fluctuatiion prediction method and server

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103049804B (zh) * 2013-01-15 2016-08-24 西安交通大学 一种基于量子力学和社交网络的股票价格趋势预测方法
US9256670B2 (en) * 2013-10-10 2016-02-09 International Business Machines Corporation Visualizing conflicts in online messages
CN105488154A (zh) * 2015-11-28 2016-04-13 小米科技有限责任公司 主题应用推荐方法及装置
CN105893611B (zh) * 2016-04-27 2020-04-07 南京邮电大学 一种构建面向社交网络的兴趣主题语义网络的方法
CN107025299B (zh) * 2017-04-24 2018-02-27 北京理工大学 一种基于加权lda主题模型的金融舆情感知方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2618651A1 (en) * 2008-02-25 2009-08-25 Gerard Voon Quantum computing h/w and s/w and artificial intelligence
CN102054116A (zh) * 2009-10-30 2011-05-11 财团法人资讯工业策进会 情绪分析方法、情绪分析系统及情绪分析装置
CN104995650A (zh) * 2011-12-27 2015-10-21 汤姆森路透社全球资源公司 用于使用源于社交媒体的数据和情绪分析来生成复合索引的方法及系统
CN104951807A (zh) * 2015-07-10 2015-09-30 沃民高新科技（北京）股份有限公司 股市情绪的确定方法和装置
US20170018033A1 (en) * 2015-07-15 2017-01-19 Foundation Of Soongsil University Industry Cooperation Stock fluctuatiion prediction method and server
CN106095777A (zh) * 2016-05-26 2016-11-09 优品财富管理有限公司 基于大数据的预测证券市场多空情绪指标方法

Also Published As

Publication number Publication date
CN107767273B (zh) 2021-08-31
CN107767273A (zh) 2018-03-06

Similar Documents

Publication Publication Date Title
WO2019047352A1 (zh) 基于社交数据的资产配置方法、电子装置及介质
CN107704512B (zh) 基于社交数据的金融产品推荐方法、电子装置及介质
JP5165033B2 (ja) 通信テキスト分類方法及び装置
WO2022141861A1 (zh) 情感分类方法、装置、电子设备及存储介质
CN103154991B (zh) 信用风险采集
Liu et al. Incrests: Towards real-time incremental short text summarization on comment streams from social network services
CN107958014B (zh) 搜索引擎
US20130282704A1 (en) Search system with query refinement
CN113934941B (zh) 一种基于多维度信息的用户推荐系统及方法
JP6056610B2 (ja) テキスト情報処理装置、テキスト情報処理方法、及びテキスト情報処理プログラム
WO2020233344A1 (zh) 一种搜索方法、装置及存储介质
WO2017091985A1 (zh) 停用词识别方法与装置
CN118172138A (zh) 一种用于电子商务的智能推荐系统
CN116644148A (zh) 关键词识别方法、装置、电子设备及存储介质
CN115269816A (zh) 基于信息处理方法的核心人员挖掘方法、装置和存储介质
CN119739838A (zh) 多标签生成匹配的rag智能问答方法、装置、设备及介质
CN113239273B (zh) 用于生成文本的方法、装置、设备以及存储介质
WO2020082609A1 (zh) 利用深度学习模型的证券研报分析方法及装置
JP6260678B2 (ja) 情報処理装置、情報処理方法、及び情報処理プログラム
CN113221538A (zh) 事件库构建方法和装置、电子设备、计算机可读介质
CN120563243A (zh) 舆情数据的处理方法、装置、设备及存储介质
CN110941713B (zh) 基于主题模型的自优化金融资讯版块分类方法
CN111222032A (zh) 舆情分析方法及相关设备
CN113704409B (zh) 一种基于级联森林的虚假招聘信息检测方法
Ferdous et al. Assortment of bangladeshi e-commerce site reviews using machine learning approaches

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17924006

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 24.09.2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17924006

Country of ref document: EP

Kind code of ref document: A1