JP5512495B2 - Data download device and data download method - Google Patents

Description

  The present invention relates to a data download apparatus and a data download method for prefetching linked data.

  When displaying a web page on a mobile terminal or the like, rendering is performed in advance for a web page (for example, a linked web page) that the user may access in the future, and the rendering result is stored in a storage area in advance. A look-ahead function is known. When a user accesses a pre-read web page, the rendering result that has been generated and stored in advance is taken out of the storage area and immediately displayed, thereby reducing the experience time required for displaying the web page, Convenience is improved for the user. In Patent Document 1, a link destination list for each web page is stored in a link table together with a selection frequency based on a past history, and URL data prefetching with a high past selection frequency is performed based on the contents of the link table. A web page prefetching method that reads a page that is highly likely to be selected into a cache area in advance is disclosed. By prefetching linked web pages that are frequently accessed to past linked pages in this way, the experience time required to display linked web pages that are likely to be accessed in the future can be reduced. .

  In addition, an information distribution service that can distribute various information to mobile terminals of service subscribers has been known, for example, i-concier (registered trademark). i-concier (registered trademark) is a service that delivers information related to a user's life as a message, such as railway operation information, road traffic information, weather information, earthquake information, event information, and coupon information. For example, a message is displayed near the service-providing character residing on the standby screen and displayed inside this balloon, or displayed on a specific date in the calendar in conjunction with the calendar of the schedule function. can do. The link destination URL may be described in the message, and the user who has viewed the message can access the link destination web page by clicking the link in order to obtain more detailed information.

Japanese Patent Laid-Open No. 11-024982

  Even if a web page is not frequently accessed by a user, there is a web page that should be prefetched as soon as possible depending on the situation. This is the railway operation information site when the schedule is greatly disturbed by weather conditions. In addition, since the update frequency is frequent and the content after the update is important, there is a web page to be prefetched immediately before the user accesses. This includes news sites. The conventional web page prefetching method can only prefetch based on the past access frequency of the user. Therefore, it is difficult to prefetch web pages that need to be accessed in an emergency, although they usually do not access much. In addition, with the conventional web page prefetching method, the prefetching timing cannot be controlled, so it is possible to prefetch web pages whose updated content is important at the optimal timing immediately before the user accesses. Absent. Therefore, according to the present invention, the prefetch timing determined according to the category to which the message belongs can be used for the link destination web page described in the message delivered by the information delivery service, so that the prefetch can be performed at the optimum timing for the user. A data download device is provided.

  The data download device of the present invention includes a message receiving unit, an operation determining unit, and a prefetching unit.

  The message receiving unit receives the distributed message and outputs the received message and the category to which the message belongs. The operation determining unit acquires the output message and category, selects a prefetch operation according to the acquired category, and outputs the selected operation and the message as a pair. The prefetching unit acquires the output action and message pair, accesses the URL described in the acquired message at the prefetching timing corresponding to the acquired operation, and prefetches the web page corresponding to the URL. Do.

  According to the data download device of the present invention, for the link destination web page described in the message distributed by the information distribution service, the pre-read timing determined according to the category to which the message belongs is used, so that the timing is optimal for the user. It becomes possible to perform prefetching.

1 is a block diagram illustrating a configuration example of a data download device according to a first embodiment. 6 is a flowchart illustrating a prefetch operation of the data download apparatus according to the first embodiment. 6 is a flowchart illustrating a prefetch timing determination operation of the data download apparatus according to the first embodiment. FIG. 3 is a diagram illustrating an operation determination table of the data download device according to the first embodiment. The figure which illustrates the message which a category extraction means makes the object of category extraction. The figure which illustrates the result of having totaled the frequency | count of access to a link destination web page per day, week, and month.

  Hereinafter, embodiments of the present invention will be described in detail. In addition, the same number is attached | subjected to the structure part which has the same function, and duplication description is abbreviate | omitted.

  First, categories handled by the data download apparatus of the present invention will be described. A category is a rough classification of messages distributed by an information distribution service. For example, if a message deals with current news, the message can be classified into the category “news”. If the content of the message is any sales promotion information, the message can be classified into the category “advertisement”. The categories are defined and assigned in advance by a distribution company that distributes messages or a service provider that provides information distribution services. It is assumed that there is a certain amount of messages that are delivered to subscribers without a category. Hereinafter, a message distributed in a state where a category is assigned is expressed as a “message whose category is known”. On the other hand, a message distributed without a category is expressed as a “message whose category is unknown”.

  Next, the operation determination table will be described in advance. As will be described in detail later, the data download apparatus according to the present embodiment is characterized in that the prefetching operation is determined for each category in accordance with the operation determined in advance by the operation determination table. This will be specifically described with reference to FIG. FIG. 4 is a diagram illustrating an operation determination table of the data download apparatus according to the present embodiment. The action determination table includes any of “immediate prefetch” operation, “history prefetch” operation, and “prefetch non-execution” in advance for each category such as “weather”, “traffic information”, “news”, “column”, “advertisement”, etc. Is set. In the example of FIG. 4, the operation set in the category “weather” is an “immediate look-ahead” operation. The operation set in the category “traffic information” is also an “immediate look-ahead” operation. For example, when a message belonging to a category such as “weather” or “traffic information” is distributed from an information distribution service, the content of the linked web page described in this message may be relatively urgent. high. Because of the nature of messages belonging to the category “weather” and category “traffic information”, it is assumed that they are often delivered when some kind of accident occurs in “weather” or “traffic information”. There are (for example, typhoon information, wave warning, train delay information, etc.). Therefore, since it is considered that such a linked web page that is likely to have some information different from normal time is often important information for the user, the operation determination table includes categories “weather”, “traffic” "Immediate look-ahead" operation is set for "Information". On the other hand, the operation set in the category “news” is the “history prefetching” operation. “History prefetch” operation refers to the user's access history to the linked web page described in the message, sets a timer immediately before the time, day of the week, and date when the access peak occurs periodically, This refers to prefetching a linked web page in a message belonging to the category simultaneously with the time indicated by the timer. As described above, the “history prefetching” operation is characterized in that the user access time is predicted from the periodicity of the past access peak, and the prefetching operation is performed immediately before that. When a message belonging to the category “news” is distributed from an information distribution service, the linked web page described in this message is likely to be a certain news site, and may be a frequently updated web page. high. Therefore, for such a web page that is updated frequently, it may be desirable to set the prefetch timing immediately before user access so that the user can always access the latest web page. Therefore, the operation determination table sets the “history prefetching” operation for the category “news”. On the other hand, the category “column” is set to “prefetching non-execution”. As for the category “column”, whether or not the user accesses depends on a hobby problem of each user. Also, most of the linked web pages introduced in “Column” are considered to be less urgent. Therefore, in the present embodiment, the operation determination table sets “prefetching non-execution” for the category “column”. In addition, for the category “advertising”, since there are not necessarily many linked web pages that are important to the user, “prefetching non-execution” may be set, but in this embodiment, the “history prefetching” operation is set. ing. For example, a user has a habit of referring to a message about a time sale sent from a nearby supermarket at 17:00 every evening and accessing a web page on which time sale information as a link destination described in this message is posted. In this case, it is optimal for the user to set a “history prefetching” operation for the link destination web page described in the message belonging to the category “advertisement”. As described above, the category “column” and the category “advertisement” often depend on the user's individual preference, and simply cannot determine the prefetching operation. Therefore, for those categories that tend to be influenced by the user's personal preference, “No prefetch” is set by default, and depending on the user's access tendency thereafter, the “history prefetch” operation or “immediate prefetch” It is good also as changing a setting operation | movement dynamically.

  Further, the categories described in FIG. 4 are merely examples, and other categories are also conceivable. If the category classification such as the above-mentioned category “column” or category “advertising” is too rough, for example, “column” is subdivided into “entertainment column” “health column” “child care column” “regional information column”, etc. You can also. “Advertisement” can be subdivided into, for example, “supermarket” and “restaurant / pub”. The setting of the operation determination table may be initially set at the time of shipment of a communication device (for example, a mobile terminal) incorporating the data download device of the present invention, or the user of the communication device may purchase the communication device. It may be set later. Further, the setting of the operation determination table may be appropriately changed according to the user's preference.

  Next, the prefetching operation of the data download apparatus according to the first embodiment will be described in detail with reference to FIGS. FIG. 1 is a block diagram illustrating a configuration example of a data download apparatus 100 according to the present embodiment. FIG. 2 is a flowchart showing the prefetching operation of the data download apparatus 100 according to this embodiment. The data download device 100 according to the present embodiment includes a message receiving unit 110, an operation determining unit 120, an operation determining table 125, a timing determining unit 130, a timer setting unit 140, and a prefetching unit 150. The message receiving unit 110 includes a message receiving unit 111 and a category extracting unit 113. The timing determination unit 130 includes an access count unit 131, an access history storage unit 132, an access peak analysis unit 133, and a timing determination unit 134.

  The operation of each component (110, 120, 140, 150) related to the prefetching operation of the data download device 100 of this embodiment will be described in detail below. In the description, it is assumed that the data download device 100 of this embodiment is built in a mobile terminal, the user of this mobile terminal is subscribed to a certain information distribution service, and the data download device 100 is subscribed to by the user. The message distributed by the information distribution service to be acquired is acquired from time to time via the communication network 91. First, the message receiving unit 111 receives a message distributed by the information distribution service (S111), and if the category to which the received message belongs is known (S112Y), the received message and the category to which the message belongs are displayed. When the category to which the received message belongs is unknown (S112N), only the received message is output to the category extracting unit 113 and the operation determining unit 120. The category extraction unit 113 acquires a message whose output category is unknown, extracts a category to which the acquired message belongs, and outputs the extracted category to the operation determination unit 120 (S113). The category extraction unit 113 may extract the category to which the message belongs, for example, by performing morphological analysis on the acquired message. Details of the operation of the category extraction unit 113 will be described later. The operation determination unit 120 acquires the output message and the category, selects one of the history prefetch operation, the immediate prefetch operation, and the prefetch non-execution according to the acquired category (S120). If it is a prefetch operation (S121Y), the history prefetch operation and the message are paired and output to the timer setting unit 140. If the selection is an immediate prefetch operation (S121N, S122Y), the immediate prefetch operation and a message are paired and output to the prefetch unit 150. On the other hand, when the selection is not pre-read (S122N), the operation determination unit 120 outputs nothing to the timer setting unit 140.

  The operation determination unit 120 refers to the operation determination table 125 described above when determining the operation, and selects any one of the history prefetch operation, the immediate prefetch operation, and the prefetch non-execution as set in the operation determination table 125. do it. The timing determination unit 130 determines the prefetch timing based on the access history to the past linked web page, and outputs the determined prefetch timing to the timer setting unit 140. Details of the operation of the timing determination unit 130 will be described later. The timer setting unit 140 acquires a pair of a history prefetch operation and a message, acquires a prefetch timing of a category to which the acquired message belongs, sets a timer based on the acquired prefetch timing (S140), Simultaneously with the elapse of the time shown, the immediate prefetch operation and the acquired message pair are output to the prefetch unit 150 (S141Y). The prefetch unit 150 acquires a pair of immediate prefetch operation and a message, accesses a URL described in the acquired message, and prefetches a web page corresponding to the URL (S150). In this way, the link destination web page described in the message distributed by the information distribution service is pre-read using the pre-read operation determined in accordance with the category to which the message belongs, so that the pre-read is performed at the optimum timing for the user. Can be performed.

  Next, the category extraction operation performed by the category extraction unit 113 will be specifically described with reference to FIG. FIG. 5 is a diagram exemplifying messages to be extracted by the category extraction unit 113. FIG. 5 specifically shows an example of a message displayed by an information distribution service such as i-concier (registered trademark) in cooperation with the schedule function. FIG. 5 illustrates a schedule function calendar, and a scene in which an economic-related breaking news message is displayed within the frame of November 2. Here, it is assumed that the category of this economic-related breaking message is unknown. If the category to which the received message belongs is unknown, the category extracting unit 113 performs morphological analysis on the received message. In the scene of FIG. 5, the title “October manufacturing industry production index breaking news” of the message is analyzed by morpheme, and is decomposed into, for example, morpheme “October” “manufacturing industry” “production” “index” “breaking news”. The category extracting means 113 is assumed to have a built-in dictionary that associates morphemes and categories. The category extraction unit 113 refers to the dictionary for each morpheme, and if the morpheme can be associated with a certain category, the score is incremented by 1 for that category. All morphemes in the message are linked to the category, and the category with the highest score is extracted as the category to which the message belongs. For example, the morpheme “breaking news” disassembled from the message title of FIG. 5 is linked to the category “news” in the dictionary provided in the category extracting means 113. Therefore, the score of the category “news” is added by one for each morpheme “breaking news”. Note that the degree of score addition may be changed between the title and the text. For example, if there is a morpheme associated with a category in the title, add 5 to the score per morpheme, and if a morpheme associated with a category exists in the message body, add 1 to the score per morpheme. Also good. In this embodiment, a message displayed in cooperation with the schedule function is described as an example of a message with an unknown category, but what can be considered as a message with an unknown category is not limited to the above. Another example is an e-mail magazine. If the user has registered in the e-mail magazine and the e-mail magazine distribution source has not assigned a category in advance, the title and body of the e-mail magazine are analyzed for morpheme and each morpheme is analyzed. By linking to the category and calculating the score, the category to which the mail magazine belongs can be extracted. Thus, even if the category of the message is unknown, the category extraction unit 113 performs morphological analysis on the content of the message, links each morpheme to the category, and calculates the score, thereby calculating the category to which the message belongs. Can be extracted.

  Next, the prefetch timing determination operation of the data download apparatus according to the first embodiment will be described in detail with reference to FIGS. FIG. 3 is a flowchart showing the prefetch timing determination operation of the data download apparatus 100 according to the present embodiment. The operation of the timing determination unit 130 related to the prefetch timing determination operation of the data download apparatus 100 according to the present embodiment will be described in detail below.

  First, the access counting means 131 counts the number of times the user has accessed the link destination URL described in the previously distributed message for each category to which the message belongs, and stores the counted number of accesses for each category. Store in the means 132 (S131). The access peak analysis unit 133 acquires the access count stored in the access history storage unit 132, analyzes the acquired access count for each category, extracts an access peak, and determines the extracted access peak as a timing determination unit. It outputs to 134 (S133). The timing determination unit 134 acquires the output access peak, and determines the read-ahead timing for each category based on whether the acquired access peak is visited in a daily period, a weekly period, a monthly period, or an arbitrary period. The determined prefetch timing is output to the timer setting unit 140 (S133a to S133d, S134a to S134d). Subsequent operations of the timer setting unit 140 and the prefetching unit 150 are as described above.

  Details of the operation performed by the access peak analyzing means 133 will be specifically described with reference to FIG. FIG. 6 is a diagram exemplifying the result of counting the number of accesses to the linked web page in units of days, weeks, and months. FIG. 6 is a result of counting the number of times a certain user accessed the link destination URL described in the message belonging to the category “news” from October 1st to 31st, 2008. Also, in this explanation, the user is a company employee and operates a mobile terminal at a company desk or the like mainly in the morning time zone before work begins and the evening time zone when work is settled, It is assumed that it is a daily routine to browse messages belonging to the category “news” distributed by the information distribution service and also browse the news site that is the link destination of the message.

  In the upper graph of FIG. 6, the horizontal axis is hour (24 hour notation), and the vertical axis is the number of accesses. For example, the number of accesses counted at 8 o'clock is a count of user accesses from 8:00:00 to 8:59:59. For example, twelve times of access are counted at 17:00, which indicates that the number of accesses (total for October) from 17:00:00 to 17:59:59 was 12 times. From this graph, users accessing the linked web pages from messages belonging to the category “News” are concentrated around 8 o'clock in the morning and around 17 o'clock in the evening, and the graph has a steep peak in any time zone. You can see that As described above, since the user is a company employee, and the daily work is to watch the news delivered in the morning and evening at the company desk, the graph has such a tendency. In the middle graph of FIG. 6, the horizontal axis represents the day of the week (Month, Tuesday, Wednesday,...), And the vertical axis represents the number of accesses. For example, the number of accesses counted on Tuesday is a count of user accesses from 00:00:00 to 23:59:59 on Tuesday. For example, seven visits are counted on Wednesday, indicating that there were seven accesses (total in October) from 0:00:00 to 23:59:59 on Wednesday. . From this graph, it can be seen that the user's access to the link destination web page from the message belonging to the category “news” is concentrated from Monday to Friday, and there is no access on Saturday or Sunday. Further, unlike the upper graph, the middle graph has no sharp peak. As described above, the graph shows such a tendency because the user is an office worker, and it is customary to watch the delivered news from Monday to Friday, which is mainly working for the company. In the lower graph of FIG. 6, the horizontal axis represents the day (1st, 2nd,..., 31st, both dates of October), and the vertical axis represents the number of accesses. For example, the number of accesses counted per day is a count of user accesses from 00:00:00 to 23:59:59 per day. For example, two accesses are counted on the 6th, which indicates that there were 2 accesses from 0:00:00 to 23:59:59 on the 6th. Also from this graph, it can be seen that the user's access to the linked web page from the message belonging to the category “news” is concentrated from Monday to Friday, and there is no access on Saturday or Sunday. Further, like the middle graph, there is no sharp peak in the lower graph. As described above, the graph shows such a tendency because the user is an office worker, and it is customary to watch the delivered news from Monday to Friday, which is mainly working for the company. As described above, the access peak analysis unit 133 obtains a monthly total result of the number of accesses on a daily, weekly, and monthly basis. The access peak analysis means 133 extracts an access peak from the graph of the tabulated result of the number of accesses for each day, week, and month. Specifically, for example, the access peak analysis unit 133 obtains the amount of change in the number of accesses for each section (every hour in the case of a daily cycle graph, one day in a week cycle, or one day in a month cycle), and the sign of the amount of change. Peaks can be detected by extracting conversion points. Alternatively, the number of accesses may be approximated by a polynomial, and the sign conversion point of the first-order differential result of this approximate expression may be extracted. The sharpness of the peak may be obtained from the amount of change in the inclination near the code conversion point of the amount of change in the number of accesses, or the point where the absolute value of the secondary differential value in the polynomial approximation described above is the largest. May be extracted as the point where the sharpest point becomes. When the point at which the peak is sharpest is present in the graph totaled by day (upper graph in FIG. 6), the access peak analysis means 133 determines that the access peak occurs in the daily cycle (S133aY), and prefetching The timing (time) is set one hour before the time when the access peak is reached (S134a). If the point at which the peak is sharpest does not exist in the graph totaled by day (upper graph in FIG. 6), the access peak analyzing means 133 determines that the access peak does not occur in the daily cycle (S133aN). ), Whether or not the point where the peak is sharpest exists in the graph (weekly graph in FIG. 6) tabulated on a weekly basis. When the point at which the peak becomes sharpest is present in the weekly graph (the graph in the middle of FIG. 6), the access peak analysis means 133 determines that the access peak occurs in the weekly cycle (S133bY), and prefetching The timing (day of the week) is set to, for example, 23:00 on the day of the week before the day of the access peak (S134b). When the point at which the peak is sharpest does not exist in the graph (weekly graph in FIG. 6) that is tabulated in the week unit, the access peak analysis unit 133 does not generate the access peak in either the daily cycle or the week cycle. Is determined (S133bN), and it is checked whether or not the point at which the peak is sharpest exists in the graph (monthly graph in FIG. 6) tabulated on a monthly basis. If the point at which the peak is sharpest is present in the graph (monthly graph in FIG. 6) that is tabulated monthly, the access peak analysis means 133 determines that the access peak occurs in the monthly cycle (S133cY), and the prefetch timing ( (Date) is set to, for example, 23: 00: 00: 00 one day before the date of the access peak (S134c). If the point at which the peak is sharpest does not exist in the graph (monthly graph in FIG. 6) that is tabulated monthly, the access peak analysis means 133 may determine whether the access peak is a daily cycle, a week cycle, or a monthly cycle. It is determined that it has not occurred (S133cN), and it is checked whether or not an access peak occurs in an arbitrary period. Here, the arbitrary period means that the access peak arrives at a period different from any of the above-mentioned daily period, weekly period, and monthly period. Specifically, there is a case where an access peak occurs at a high frequency such as once every 20 minutes, or a case where an access peak occurs at a low frequency such as once every four months. The access peak analyzing means 133 can generate the above-mentioned access aggregation graph in units of minutes, years, etc., for example, and can extract access peaks in an arbitrary period from the generated graph as described above. If the point at which the peak becomes sharp is present in the graph totaled in minutes, years, etc., the access peak analysis means 133 determines that the access peak occurs in an arbitrary period (S133dY), and determines the prefetch timing as the access peak. (S134d). When the peak does not exist in the graph totaled in minutes, years, etc., the access peak analysis unit 133 does not generate the access peak in any of the daily cycle, the week cycle, and the monthly cycle. It is determined that no period has occurred (S133dN), and the process is terminated. When the access peak analyzing unit 133 finishes the process through step 133dN, no output is performed to the timing determining unit 134. Therefore, in this case, the timing determination unit 134, the timer setting unit 140, and the prefetch unit 150 do not perform any processing, and the prefetch operation is not performed.

  In the example of FIG. 6, the access peak analysis unit 133 determines that an access peak occurs every day at 8:00 am (daily cycle) (S133aY). The timing determination unit 134 sets the prefetch timing (time) to 7:00 that is one hour before 8:00, which is the time when the access peak occurs (S134a). Since the total number of accesses at the time of access peak is the total number of accesses from 8:00:00 to 8:59:59, the time when the access peak is set to 8:30:00, one hour before that It is good also as 7:30:00. Further, the prefetch timing does not necessarily have to be one hour before the access peak, and may be, for example, 30 minutes before. Further, the access peak is not necessarily limited to only one place. For example, in the upper graph of FIG. 6, the sharpest access peak is visited at 8 am, but the peak visited at 17:00 is the second sharpest, and the peak sharpness is the peak at 8 am and the peak at 17:00 in the evening. There is no big difference. Therefore, the number of access peaks is not limited to one, but, for example, the top N ranks (N is an arbitrary natural number of 2 or more) may be selected, and a plurality of prefetch timings may be set from each access peak. In this way, the access peak analyzing unit 133 extracts the access peak, and the timing determining unit 134 is based on whether the extracted access peak visits in a daily cycle, a week cycle, a monthly cycle, or an arbitrary cycle. Since the prefetch timing is determined for each category, it is possible to perform prefetching in consideration of the tendency of the user to access the linked webpage, and even if it is a linked webpage of a category that frequently updates, The latest information that the user wants to browse can be prefetched.

Claims (12)

A message receiving unit that receives the distributed message and outputs the received message and a category to which the message belongs;
An operation determining unit that acquires the output message and category, selects a prefetch operation according to the acquired category, and outputs the selected operation and the message as a pair;
The pair of the output operation and message is acquired, and the URL described in the acquired message is accessed at the prefetch timing corresponding to the acquired operation, and the web page corresponding to the URL is prefetched. A data download device comprising a prefetch unit. A message receiving unit that receives the distributed message and outputs the received message and a category to which the message belongs;
Acquire the output message and category, select one of the history prefetch operation, immediate prefetch operation, prefetch non-execution according to the acquired category, if the selection is an operation, An operation determining unit for outputting the selected operation and the message as a pair;
The access history to the link destination URL described in the message delivered in the past is acquired, the prefetch timing is determined for each category to which the message belongs based on the acquired access history, and the determined prefetch timing A timing determination unit for outputting
When a pair of the operation and message is acquired, and the acquired operation is a history prefetch operation, the prefetch timing of the category to which the acquired message belongs is obtained, and a timer is set based on the acquired prefetch timing. A timer setting unit that outputs a pair of the acquired prefetch operation and the acquired message simultaneously with the elapse of the time indicated by the timer;
When the output operation and message pair is acquired and the acquired operation is an immediate prefetch operation, the URL described in the acquired message is accessed, and the prefetching of the web page corresponding to the URL is performed. A data download device comprising: a prefetching unit that performs the operation. The data download device according to claim 2,
The timing determination unit
An access count means for counting the number of accesses to the link destination URL described in a message distributed in the past for each category to which the message belongs, and outputting the counted number of accesses;
Obtaining the output number of accesses, extracting the access peak by analyzing the acquired number of accesses, and outputting the extracted access peak;
A data download device comprising: timing determination means for acquiring the output access peak and determining a prefetch timing for each category based on the acquired access peak. The data download device according to claim 3,
The timing determination means;
A data download apparatus, wherein the prefetch timing is determined based on whether the acquired access peak is visited in a daily cycle, a week cycle, or a monthly cycle. The data download device according to any one of claims 1 to 4,
The message receiver
When the received message is received and the category to which the received message belongs is known, the received message and the category to which the message belongs are output, and the category to which the received message belongs is unknown A message receiving means for outputting only the received message;
Category extraction means for acquiring the output message and extracting the category to which the acquired message belongs if the category to which the acquired message belongs is unknown, and outputting the extracted category. A data download device characterized by that. The data download device according to claim 5,
The category extracting means includes
A data download apparatus, wherein a category to which the acquired message belongs is extracted by morphological analysis. A message receiving step of receiving the distributed message and outputting the received message and a category to which the message belongs;
An operation determining step of acquiring the output message and category, selecting a prefetch operation according to the acquired category, and outputting the selected operation and the message as a pair;
The pair of the output operation and message is acquired, and the URL described in the acquired message is accessed at the prefetch timing corresponding to the acquired operation, and the web page corresponding to the URL is prefetched. A data download method comprising a prefetch step. A message receiving step of receiving the distributed message and outputting the received message and a category to which the message belongs;
Acquire the output message and category, select one of the history prefetch operation, immediate prefetch operation, prefetch non-execution according to the acquired category, if the selection is an operation, An operation determining step for outputting the selected operation and the message as a pair; and
The access history to the link destination URL described in the message delivered in the past is acquired, the prefetch timing is determined for each category to which the message belongs based on the acquired access history, and the determined prefetch timing A timing determination step for outputting
When a pair of the operation and message is acquired, and the acquired operation is a history prefetch operation, the prefetch timing of the category to which the acquired message belongs is obtained, and a timer is set based on the acquired prefetch timing. A timer setting step for outputting an immediate prefetch operation and the acquired message pair at the same time as the time indicated by the timer;
When the output operation and message pair is acquired and the acquired operation is an immediate prefetch operation, the URL described in the acquired message is accessed, and the prefetching of the web page corresponding to the URL is performed. A data download method comprising: a pre-reading step. The data download method according to claim 8,
The timing determination step comprises:
An access count substep for counting the number of accesses to the link destination URL described in a message distributed in the past for each category to which the message belongs, and outputting the counted number of accesses;
Obtaining the output number of times of access, analyzing the acquired number of times of access to extract an access peak, and outputting the extracted access peak, an access peak analysis sub-step;
A data download method comprising: a timing determination sub-step of acquiring the output access peak and determining a prefetch timing for each category based on the acquired access peak. The data download method according to claim 9,
The timing determination substep comprises:
A data download method, wherein the prefetch timing is determined based on whether the acquired access peak is visited in a daily cycle, a week cycle, or a monthly cycle. The data download method according to any one of claims 7 to 10,
The message receiving step comprises:
When the received message is received and the category to which the received message belongs is known, the received message and the category to which the message belongs are output, and the category to which the received message belongs is unknown Includes a message reception sub-step for outputting only the received message, and
When the acquired message is acquired and the category to which the acquired message belongs is unknown, the category to which the acquired message belongs is extracted, and a category extraction sub-step for outputting the extracted category; A data download method comprising: a data download method comprising: The data download method according to claim 11, comprising:
The category extraction substep includes:
A data download method, wherein a category to which the acquired message belongs is extracted by morphological analysis.

Images