JP2019095958A - Market condition comment generation assisting device and market condition comment generation assisting method - Google Patents

Abstract

To increase the efficiency of creating a market condition comment of appropriate quality to reduce the work load of an analyst, etc.SOLUTION: A market condition comment generation assisting device 100 is configured to include: a storage device 101 for storing text based news report data 125 obtained from a first source and numeric value based market condition data 126 obtained from a second source; and an arithmetic unit 104 for executing a process for applying a prescribed sentence analysis algorithm to the news report data and specifying an important sentence whose influence degree regarding market conditions is higher than or equal to a prescribed standard among the sentences included in the news report data, a process for selecting one important sentence on the basis of either the expression of description regarding a prescribed event being stronger in each of the important sentences or the fluctuation width of the market condition data reckoning from a news report day being larger in each of the important sentences, and specifying the news report day of the selected important sentence as an important day, and a process for combining a sentence derived by applying the market condition data on the important day to a prescribed market condition comment template and the important sentence of the important day and thereby generating a market condition comment.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a market comment generation support apparatus and a market comment generation support method, and more particularly, to a technology capable of efficiently creating a market comment of appropriate quality and reducing the workload of an analyst etc. .

  A financial institution performs investment advice to a client, or manages investment trust based on a fund deposited by the client, in addition to the self funded investment. In addition, financial institutions that carry out such investment related business regularly issue market reports, for example, to customers.

  The above-mentioned market report includes a text called market comment that describes the reason or effect of market fluctuation. The comment on the market should be prepared by referring to the market value of the target period, the fact that the market affected the news (news article), the influence of the market fluctuation (news article), etc. Is common.

  As a prior art related to information provision accompanying such investment work etc., for example, dictionary data storage for storing dictionary data that defines the correspondence between a specific expression character string and an abstraction tag indicating the type thereof Means, stock data storage means for storing digitized article data, combination of abstract tags included in article data, or combination of abstract tags and a specific character string, stock price fluctuation event information The extraction rule storage means for storing the extraction rule that defines the character string to be extracted as the component of and the column of at least the brand and the event type when the request for stock price fluctuation event search is transmitted from the client terminal Means for sending a search screen provided with the symbol and prompting for designation of brand and event type, and from the client terminal, brand and event type When specification information is sent, the article data storage means is searched using the brand name and event type as keys, and article data extraction means for extracting related article data, and morphemes for decomposing the extracted article data into morphemes The analysis means, the abstracting means for giving the corresponding abstract tag to the character string representing at least the brand name, the event type, and the time in each morpheme by referring to the above dictionary data, and the above extraction for each article data Event information generating means for extracting stock price fluctuation event information by extracting a character string representing at least a stock, a character string representing an event type, and a character string representing an event occurrence time by applying a rule, and the stock price fluctuation event information Means for generating a search result screen in which the information is described and transmitting the screen to the client terminal. Etc. (see Patent Document 1) has been proposed.

JP, 2011-204199, A

  There is not much burden on analysts, etc. in making the above-mentioned market comment. For example, it takes a lot of man-hours in the task of creating documents while referring to market value figures and news reports that have huge numbers and variations and utilizing them on their own understanding.

In addition, the tangible and intangible experiential knowledge that analysts and others should prepare for the business is a strong belonging. Therefore, differences in the analysts' skills and experience are likely to cause large variations in the efficiency and quality of the market comment that is the product.
Therefore, an object of the present invention is to provide a technology that makes it possible to streamline the preparation of market comment of appropriate quality and reduce the workload of analysts and the like.

  A market comment generation support device according to the present invention for solving the above problems includes: a storage device for storing text-based news report data obtained from a first source; and numerical-based market data obtained from a second source; A predetermined sentence analysis algorithm is applied to the report data, and a process of specifying important sentences having a degree of influence on the market among the sentences included in the report data, and a description related to a predetermined event in each of the important sentences One important sentence is selected on the basis of one having a stronger expression, or one having a larger fluctuation range of the market data starting from the day of coverage in each of the important sentences, and selecting the one important sentence A process that identifies the date of news report as an important date, the text that applies the market condition data of the important day to a predetermined market comment form, and the important sentence of the relevant important day Characterized in that it comprises an arithmetic unit that executes a process of generating a comment, and the.

  Further, the market comment generation support method of the present invention is an information processing apparatus including a storage device for storing text-based news report data obtained from a first source and numerical-based market data obtained from a second source. The apparatus applies a predetermined sentence analysis algorithm to the news report data, and among the sentences included in the news report data, identifies an important sentence having a predetermined impact on market conditions and a predetermined event in each of the important sentences Select one important sentence on the basis of either the one with a stronger description of the statement or the one with a greater fluctuation in the market data starting from the day of coverage in each of the important sentences, A process to identify the news report date of the selected important sentence as the important day, and a sentence obtained by applying the market data of the important day to a predetermined market commentary pattern and the important sentence of the important day And executes a processing for generating the market comments and.

  According to the present invention, it is possible to streamline the creation of market comment of appropriate quality, and reduce the burden on the analyst and the like.

It is a network block diagram containing the market comment generation support device of this embodiment. It is a figure showing the example of hardware constitutions of the market comment generation support device in this embodiment. It is a figure which shows the data structural example of the news report data table in this embodiment. It is a figure which shows the data structural example of the market data table in this embodiment. It is a figure which shows the data structural example of the important sentence position data table in this embodiment. It is a figure which shows the data structural example of the important word data table in this embodiment. It is a figure which shows the example of a data structure of the description score table in this embodiment. It is a figure which shows the example of a data structure of the description score table in this embodiment. It is a figure which shows the example of a main flow of the market value comment production | generation assistance method in this embodiment. It is a figure which shows the conceptual example of the market value comment production | generation support method in this embodiment. It is a figure which shows the example 1 of a flow of the market value comment production | generation assistance method in this embodiment. It is a figure which shows Screen example 1 in this embodiment. It is a figure which shows the example 2 of a flow of the market value comment production | generation assistance method in this embodiment. It is a figure which shows the example 3 of a flow of the market value comment production | generation support method in this embodiment. It is a figure which shows the example 4 of a flow of the market value comment production | generation support method in this embodiment. It is a figure which shows the example 2 of a screen in this embodiment. It is a figure which shows Screen example 3 in this embodiment. It is a figure which shows Screen example 4 in this embodiment. It is a figure which shows the example 5 of a flow of the market value comment production | generation support method in this embodiment. It is a figure which shows the example of the number of articles classified by topic in this embodiment. It is a figure which shows the example of the topic temperature in this embodiment.

---Network configuration example---

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a network configuration diagram including the market comment generation support device 100 according to the present embodiment. The market condition comment generation support device 100 shown in FIG. 1 is a computer device that streamlines the creation of market condition comments of appropriate quality and can reduce the workload of an analyst or the like.

  The market comment generation support device 100 according to the present embodiment is communicably connected to the user terminal 200, the news report distribution server 300, and the market data distribution server 400 via an appropriate network 10 such as the Internet or a corporate network. There is.

  Among them, the user terminal 200 is a terminal operated by a person in charge of creating a market comment, such as an analyst of a financial institution or a fund manager. As this user terminal 200, a general PC or tablet terminal, a smart phone, etc. can be assumed.

  Further, the news report distribution server 300 is a server device operated by a news report source such as a newspaper company or a communication company that provides a news report distribution service, and as described above, the market value comment generation support device 100 of the present embodiment via the network 10. And, for example, delivers text-based news report data at predetermined time intervals.

Further, the market data distribution server 400 is a server device operated by a news report source such as a newspaper company or a communication company that provides a market data distribution service, and the market comment generation support of the present embodiment is performed via the network 10 as described above. It is connected to the apparatus 100, and delivers, for example, numerical-based market data at predetermined time intervals.
--- Hardware configuration ---

  The hardware configuration of the market comment generation support device 100 in the present embodiment is as follows. FIG. 2 is a diagram showing an example of a hardware configuration of the market value comment generation support device 100 in the present embodiment.

  The market comment generation support device 100 includes a storage device 101 configured by an appropriate non-volatile storage element such as a solid state drive (SSD) or a hard disk drive, a memory 103 configured by a volatile storage element such as a RAM, and the storage device 101. The stored program 102 is read out to the memory 103 and executed to perform overall control of the apparatus itself, as well as connected to the computing apparatus 104 such as a CPU that performs various determinations, computations and control processing, and the network 10. A news distribution server 300, and a communication device 105 responsible for communication processing with other devices such as a market data distribution server 400 are provided.

In addition to the program 102 for implementing the functions necessary for the market comment generation support device 100 of the present embodiment, the storage device 101 includes a news report data table 125, a market data data table B 126, an important sentence position table 127, At least the important word table 128 and the descriptive score table 129 are stored. The details of each of these tables will be described later.
--- Data structure example ---

  Subsequently, tables used by the market value comment generation support device 100 of the present embodiment will be described. FIG. 3A shows an example of the data configuration of the news report data table 125 in the present embodiment.

  The news report data table 125 is a table storing, for example, text-based news report data obtained at predetermined time intervals from the news report distribution server 300 that provides a news report distribution service.

  The data structure is a collection of records consisting of an article title, an article content, and a delivery date and time, which clearly show the coverage data, with the source name uniquely identifying the source reporting company as a key.

  Further, FIG. 3B shows a data configuration example of the market data table 126 in the present embodiment. The market data table 126 is a table storing numerical market data obtained from the market distribution server 400 that provides the market data distribution service.

  The data structure is a collection of records consisting of data such as target market, various market values such as open and close prices, and distribution date and time, with the source name uniquely identifying the distributor of market data as the source. It is.

  Further, FIG. 4A shows a data configuration example of the important sentence position data table 127 in the present embodiment. The important sentence position data table 127 is a table storing information on positions where important sentences are preferentially arranged for each article in each source such as a newspaper.

  The data structure is based on the title of the article, the position of the important sentence in the article, and the important sentence in the article, with the source name uniquely identifying the source of the news such as a newspaper company or a news agency. It is a collection of records consisting of data such as an unknown position (ie, the position of a sentence to be excluded).

  Further, FIG. 4B shows a data configuration example of the important word data table 128 in the present embodiment. The important word data table 128 is a table that stores keywords related to events that are likely to affect market fluctuation, which are keywords that can be included in a sentence included in the news report data. The data structure is a collection of keywords corresponding to key words.

5A shows an example of the data configuration of the description score table 129 in the present embodiment. The descriptive score table 129 is a table storing scores defined in accordance with the degree of influence on market conditions of each description in a sentence included in the news report data.
The data structure is a collection of records consisting of a score value and a description that is worthy of the score and describes the market price fluctuation situation in market conditions.

Further, in the description score table 129 illustrated in FIG. 5B, unlike the example of FIG. 5A described above, a collection of records including data such as term categories, target countries / organizations, and example terms for score values. It has become.

The descriptive score table 129 shown in FIG. 5B defines the score when the descriptive content indicated by the term example is included for the term category included in the relevant description and its subject (country or institution). Of course, the description score table 128 of FIG. 5A and the description score table 128 of FIG. 5B may be used in combination.
--- Example of flow ---

  Hereinafter, the actual procedure of the market value comment generation support method in the present embodiment will be described based on the drawings. Various operations corresponding to the market comment generation support method described below are realized by a program that the market comment generation support device 100 reads out to a memory or the like and executes. The program is composed of code for performing various operations described below.

  FIG. 6 is a diagram showing an example of a main flow of the market value comment generation support method according to the present embodiment. Moreover, FIG. 7 is a figure which shows the example of a concept of the market value comment production | generation assistance method in this embodiment.

  Here, first, an overview of the market comment generation support method will be described. In this case, the market comment generation support device 100 acquires various news reports such as market conditions and political conditions from the news distribution server 300 via the network 10, and stores the data in the news data table 125 (s100). The coverage data acquired here is text based.

  Subsequently, the market comment generation support device 100 acquires market data in various markets such as a stock market, a bond market, and a futures market from the market data distribution server 400 via the network 10, and this is stored in the market data table 126. Store (s200). The market data acquired here is on a numerical basis.

  Next, the market comment generation support device 100 receives an instruction of a predetermined person in charge of a financial institution such as an analyst or a fund manager from the user terminal 200 (s250). The instruction received here is each designation of a target period for generating a market comment and a target product (for example, a stock name of a car, a precision machine, etc.).

  Subsequently, the market comment generation support device 100 executes extraction of important sentences (s300) from among the text-based news report data obtained in s100 and s200 and the numerical base that has been designated in s250. Details of this important sentence extraction process will be described later.

  In addition, the market comment generation support device 100 selects the news report date of the important sentence extracted in s300 described above as the important day (s400). The details of the selection process of this important date will also be described later.

  Subsequently, the market comment generation support device 100 generates a description comment by setting the numerical market data in a predetermined form to generate a descriptive sentence, and combining this descriptive sentence with the important sentence extracted in s300. To do (s500).

Next, the market comment generation support device 100 extracts text-based news data and numerical-based market data, and extracts the target period designated in s250 with respect to the user terminal 200 such as the above-mentioned analyst or fund manager. Distribute a screen to list the important sentences and the generated market comment, and on this screen, an instruction regarding addition or deletion of the date used for market comment, that is, an important day, or the important day selected in the instruction The original text of the news report data is displayed, and on the screen, an instruction to add or delete an important sentence to be used for the market comment is received (s600).

  Further, when the market comment generation support device 100 receives a correction instruction from the above-described user terminal 200 in response to the market comment generated and presented in s500 or the market comment re-generated and presented in s600, the market instruction is accordingly made. The comment is corrected, this is distributed to the user terminal 200 (s700), and the process is ended.

  Next, the details of steps corresponding to the main processing among the steps shown in the above-mentioned main flow example will be described. FIG. 8 is a diagram showing a flow example 1 of the market value comment generation supporting method in the present embodiment, and more specifically shows the details of the important sentence extraction (s300).

  Here, the market comment generation support device 100 extracts, from the news data table 125, news report data of a target period which has been designated in advance from the user terminal 200 (s310). An example of the screen for receiving the specification here is as shown in the screen 900 of FIG. This screen 900 is provided with an interface for receiving designations such as keywords, sources such as news, and target periods regarding products etc. which analysts, fund managers and the like are working on regarding each market in Japan and abroad.

Next, the market comment generation support device 100 applies a predetermined text analysis algorithm (an existing one may be appropriately adopted) to the news report data obtained in s310, and the news report data is disassembled for each sentence. Each sentence obtained by the decomposition is decomposed into words (s322).
In addition, the market comment generation support device 100 executes, for each sentence, the statistics of the appearance frequency of the word obtained for each sentence in s322 (s323).

  Subsequently, the market comment generation support device 100 collates the statistical analysis result of the appearance frequency of the word obtained in s323 with the important word table 128 to calculate the appearance frequency of the important word (s331), and this appearance frequency The evaluation score according to the size of is calculated as the importance of the corresponding sentence (s333). It is assumed that the market condition comment generation supporting device 100 holds in advance the rule of the evaluation score of the importance, which is determined according to the size of the appearance frequency of the key word.

In addition, the market comment generation support device 100 refers to the important sentence position table 127 for each sentence whose importance is calculated in s333, and is arranged at a predetermined position in the news data defined in the important sentence position table 127. If yes, a predetermined evaluation score is added, and then, among sentences, one having an evaluation score equal to or higher than a predetermined standard, ie, one having an importance degree equal to or higher than the predetermined standard is selected as an important sentence (s340).
Further, FIG. 10 shows a flow example 2 of the market value comment generation support method in the present embodiment, and specifically shows the selection processing of the important date.

  Here, the market comment generation support device 100 collates each of the important sentences selected in the above-described s300 with the description score table 128, and responds to the presence or absence of the description corresponding to the description regarding each event defined in the description score table 128. Then, the evaluation score of the important sentence is calculated (s410: s411 to s413).

In addition, the market comment generation support device 100 acquires market data on s200 starting from the news release date of each of the above-mentioned important sentences (if necessary, data of the corresponding period is separately acquired from the market data table 126) And the fluctuation range of the value (example: closing price on the day of coverage-closing price on the previous day) is calculated, and the evaluation score according to the size of this fluctuation range is specified (s420: s42
1 to s423). Of course, it is assumed that the market condition comment generation supporting device 100 holds in advance the rule of the value of the evaluation score according to the magnitude of the above-mentioned fluctuation range in the program 102.

  Subsequently, the market comment generation support device 100 selects one important sentence based on the evaluation score calculated in s 410 and s 420 respectively, and selects the important day, specifying the news coverage date of the selected important sentence as the important day. A process is performed (s430).

It is assumed that the important sentence selection rule, ie, the selection rule, is received in advance from the user terminal 200. In addition, as a specific example of such a selection rule, for example, it is possible to assume one that selects the day on which the absolute value of the evaluation score of the important sentence obtained in s410 is the largest. Alternatively, it can be assumed to select a day in which the absolute value of the evaluation score according to the magnitude of the fluctuation range obtained in s420 is the largest. It is also conceivable to select the day on which the absolute value of the evaluation score of the important sentence obtained in s410 is the maximum and the absolute value of the evaluation score according to the size of the fluctuation range obtained in s420 is the maximum. Alternatively, it may be assumed to select a day on which the absolute value of the evaluation score is the largest according to the magnitude of the fluctuation range.
Further, FIG. 11 shows a flow example 3 of the market value comment generation support method according to the present embodiment, and more specifically shows generation processing of the market value comment.

  Here, the market comment generation support device 100 applies the market date data (for example, the opening price and the closing price obtained in s200, for example, the opening price and the closing price) of the important day specified in s400 to the predetermined market comment form. A market comment is generated by generating a sentence (s510) and combining it with the important sentence of the important day (the important day obtained in s400) (s520).

  FIG. 12 shows a flow example 4 of the market value comment generation support method according to the present embodiment, and more specifically shows the process of accepting selection date / confirmation confirmation / correction.

  Here, the market comment generation support device 100 causes the user terminal 200 to distribute and display the market comment generated in the above-described s520 and the information related thereto (report data of the target period, market data, important sentences, etc.) Then (s610), the user receives an important date desired by the user during the market comment period, that is, an instruction for specifying or deleting an important sentence corresponding to the important date (s620). With regard to such an important sentence selection operation, an analyst, a fund manager, or the like selects the relevant article (important sentence) in the "article summary of the target period" column 1210 in the screen 1200 of FIG. As a result, it is illustrated that important sentences are listed in the sentence group column 1220 of the selected article (details are the screen 1400 of FIG. 15).

In addition, the market comment generation support device 100 distributes to the user terminal 200 information on the important sentences of the important day designated in s620 (selected article sentence group column 1220 and focused article full text column 1215 in the screen 1200 of FIG. 13). (S630), and the market comment generation (S640) is performed again based on the instruction received in S620. In the screen 1200 of FIG. 13, an example generated in this way is shown as a market comment 1230.
FIG. 16 shows a flow example 5 of the market value comment generation support method according to the present embodiment, and more specifically shows the process of calculating the topic temperature.

  Here, the market comment generation support device 100 executes score evaluation according to a predetermined rule with respect to the continuation tendency of the news about a predetermined topic included in the sentence, regarding each sentence included in the news data, when specifying the above-mentioned important sentence.

In this case, the market comment generation support device 100 generates the important word table 128 in each sentence (already extracted in the decomposition in s321) in the news data already obtained in s100.
The topic of the sentence is specified by determining whether a predetermined keyword defined in an appropriate dictionary or the like is included (s310, s351).

  Subsequently, the market comment generation support device 100 counts, as an initial value, the number of appearances of the above-mentioned keyword on the news report day of the sentence and a predetermined attenuation rate (e.g. 0.5) along with the number of days elapsed from the news report day. The temperature of the topic is calculated according to a predetermined algorithm of decaying with / (day) and rising according to the number of new appearances of the corresponding keyword after the date of coverage (s 352).

  Among the above, an example of the number of articles by topic 1700 where the number of appearances of keywords is counted for each day is shown in FIG. As shown in this example, for example, the topic “UK-EU leaving” is again set to “1” for the number of corresponding keywords on the third day from the date of coverage, and the keywords on the 23rd, 24th, and 29th days thereafter. It can be seen that “1” each appeared in the news data.

  Further, when the topic temperature is calculated based on the example of the number of articles by topic 1700, as shown in FIG. 18, for example, the topic temperature of the topic "UK-EU departure" has an initial value of "1" for the first day of coverage. On the second day, it can be calculated that 1 × 0.5 + 1 (one incident in the news data on the second day) = 1.5 degrees, and so on.

  Subsequently, the market comment creation support device 100 identifies an topic whose topic temperature calculated in s352 exceeds a predetermined value and continues for a predetermined period as an important topic (s 353). Score evaluation of the corresponding sentence is performed according to the number, and is reflected in the selection of important sentences (s 340). Specifically, the evaluation score is added to the calculation result of s 338 in the flow of FIG.

As mentioned above, although the best mode etc. for carrying out the present invention were concretely explained, the present invention is not limited to this, and can be variously changed in the range which does not deviate from the gist.
According to the present embodiment, it is possible to streamline the creation of market comment of appropriate quality, and reduce the workload of the analyst etc.

  At least the following matters will be made clear by the description of the present specification. That is, in the market value comment generation support device of the present embodiment, the computing device is disposed at a predetermined position in news report data which is known in advance for each source with respect to each of the sentences when specifying the important sentences. Score evaluation is performed according to a predetermined rule regarding at least one of the number of appearances of a predetermined key word and / or the continuation tendency of the news concerning a predetermined topic included in the sentence, and the sentence with the highest score is important It may be specified as a sentence.

  According to this, it becomes possible to identify precisely what is worthy of adoption as a market condition comment from each sentence including news data, without imposing a special burden on the analyst etc., with high accuracy. As a result, it is possible to further streamline the preparation of market comment of suitable quality and reduce the burden on the analyst and the like.

  Further, in the market value comment generation supporting device of the present embodiment, the computing device is configured to calculate the score of the continuation tendency of the news report on the topic with respect to the predetermined topic indicated by the predetermined keyword in each of the sentences. The topic's temperature is calculated according to a predetermined algorithm that decays at a predetermined rate with the passage of time with the number of appearances as the initial value, and increases according to the number of new appearances of the keyword after the news coverage date. It is also possible to specify an topic that has continued on a day beyond the predetermined period as an important topic, and score evaluation of the corresponding sentence is performed according to the presence and the number of the important topic.

  According to this, from each sentence including news report data, it relates to the event that news was continuously reported and deserves to be adopted more as a market comment, without imposing a special burden on analysts etc. It becomes possible to identify well. As a result, it is possible to further streamline the preparation of market comment of suitable quality and reduce the burden on the analyst and the like.

  Further, in the market value comment generation support device of the present embodiment, the arithmetic device generates the market price comment generated, news report data and market data of the storage device, and a sentence specified as an important sentence when the market comment is generated. Outputting to a predetermined device, accepting a correction instruction of at least one of the important sentence and the important date by a predetermined user, and re-specifying at least one of the important sentence and the important day according to the correction instruction; It is also possible to further execute a process of generating a market condition comment again.

  According to this, analysts etc. will be clearly presented from the press data about what was used to identify important sentences and important days, and it will be easier to judge market comment or correction in the analyst etc. It can lead to the generation of market comment with good accuracy. As a result, it is possible to further streamline the creation of market comment of better quality, and as a result, it is possible to reduce the workload of analysts and the like.

  Further, in the market value comment generation support method of the present embodiment, the information processing apparatus is disposed at a predetermined position in news report data which is previously known for each source with respect to each of the sentences when specifying the important sentences. Score evaluation is performed according to a predetermined rule regarding at least one of the number of appearances of important words and whether the number of appearances of predetermined key words and / or the continuation tendency of the news regarding the predetermined topic included in the sentence It may be specified as an important sentence.

  Further, in the market value comment generation support method according to the present embodiment, the information processing apparatus performs the score evaluation on the continuation tendency of the news report on the topic, the keyword at the news report date regarding the predetermined topic indicated by the predetermined keyword in each of the sentences. The topic's temperature is calculated according to a predetermined algorithm that decays at a predetermined rate with the passage of time, with the number of appearances of the topic as an initial value, and increases according to the number of new appearances of the keyword after the date of coverage. It is also possible to identify an important topic that has a date that is equal to or greater than the value and continue beyond a predetermined period as an important topic, and score evaluation of the corresponding sentence according to the presence and the number of the relevant important topic.

  In the market value comment generation support method of the present embodiment, the information processing apparatus generates the market comment, the news report data and market data of the storage device, and a sentence specified as an important sentence when the market comment is generated. Is output to a predetermined device, and a correction instruction of at least one of the important sentence and the important date by a predetermined user is received, and at least one of the important sentence and the important date is again specified according to the correction instruction. The process of generating a market comment may be further executed.

10 Network 100 Market Value Comment Generation Support Device 101 Storage Device 102 Program 103 Memory 104 Arithmetic Unit 105 Communication Device 125 News Data Data Table 126 Market Data Data Table 127 Important Sentence Position Table 128 Key Word Table 129 Description Score Table 200 User Terminal 201 Memory Device 202 Program 203 memory 204 arithmetic unit 205 input unit 206 output unit 207 communication unit 300 news distribution server 400 market data distribution server

Claims (8)

A storage device for storing text-based press data obtained from the first source and numerical-based market data obtained from the second source;
A predetermined sentence analysis algorithm is applied to the report data, and a process of specifying important sentences having a degree of influence on the market among the sentences included in the report data, and a description related to a predetermined event in each of the important sentences One important sentence is selected on the basis of one having a stronger expression, or one having a larger fluctuation range of the market data starting from the day of coverage in each of the important sentences, and selecting the one important sentence A process of specifying a news report date as an important date, a process of generating a market comment by combining a sentence obtained by applying market data on the important day to a predetermined market comment form and an important sentence on the important day, A computing device to execute
The market comment generation support device characterized by having. The arithmetic device is
In identifying the important sentences, whether or not each sentence is placed at a predetermined position in the news report data previously known for each source, the number of appearances of predetermined important words, and the sentences The score evaluation is performed according to a predetermined rule regarding at least one of the continuation tendency of the coverage of the predetermined topic including, and the sentence having the highest score is specified as the important sentence.
The market comment generation support device according to claim 1, wherein The arithmetic device is
In the score evaluation of the continuation tendency of the news about the topic, regarding the predetermined topic which the predetermined keyword shows in each of the sentences, the number of appearances of the keyword on the news day as an initial value is attenuated at a predetermined rate with lapse of time Calculate the temperature of the topic according to a predetermined algorithm that rises according to the number of new appearances of the keyword after the day, and identify the topic whose days above the predetermined value continue beyond the predetermined period as important topics According to the presence and the number of the important topics, score evaluation of the corresponding sentences is performed.
The market comment generation support device according to claim 2, wherein The arithmetic device is
The generated market comment, news report data and market data of the storage device, and a sentence specified as an important sentence at the time of the market comment generation are output to a predetermined device, and at least the important sentence and the important day by a predetermined user It accepts any correction instruction, executes again the specification of at least one of the important sentence and the important date according to the correction instruction, and further executes a process of generating a market condition comment again.
The market comment generation support device according to claim 1, wherein An information processing apparatus comprising a storage device for storing text-based news report data obtained from a first source and numerical-based market data obtained from a second source;
Applying a predetermined sentence analysis algorithm to the news report data, and processing, among sentences included in the news report data, identifying important sentences having a degree of influence on market conditions equal to or higher than a predetermined standard;
In each of the important sentences, either the one with a stronger description of a given event or the one with a larger fluctuation range of the market data starting from the coverage date in each of the important sentences Processing for selecting one important sentence and specifying the date of coverage of the selected important sentence as the important day,
A process of generating a market comment by combining a sentence obtained by applying market data on the important day to a predetermined market comment form and a key sentence on the important day;
The market comment generation support method characterized by carrying out. The information processing apparatus
In identifying the important sentences, whether or not each sentence is placed at a predetermined position in the news report data previously known for each source, the number of appearances of predetermined important words, and the sentences Perform score evaluation according to a predetermined rule regarding at least one of the continuation tendency of the report regarding the predetermined topic including, and identifying the sentence with the highest score as the important sentence,
The market comment generation support method according to claim 5, characterized in that: The information processing apparatus
In the score evaluation of the continuation tendency of the news about the topic, regarding the predetermined topic which the predetermined keyword shows in each of the sentences, the number of appearances of the keyword on the news day as an initial value is attenuated at a predetermined rate with lapse of time Calculate the temperature of the topic according to a predetermined algorithm that rises according to the number of new appearances of the keyword after the day, and identify the topic whose days above the predetermined value continue beyond the predetermined period as important topics Evaluate the score of the relevant sentences according to the presence and the number of important topics
The market comment generation support method according to claim 6, characterized in that: The information processing apparatus
The generated market comment, news report data and market data of the storage device, and a sentence specified as an important sentence at the time of the market comment generation are output to a predetermined device, and at least the important sentence and the important day by a predetermined user Accept any correction instruction, execute again specification of the important sentence and / or the important date according to the correction instruction, and further execute a process of generating a market condition comment again.
The market comment generation support method according to claim 5, characterized in that:

Images