JP5545186B2 - Point search device, point search method, and point search program - Google Patents

Description

  The present invention relates to a point search device, a point search method, and a point search program that search for point candidates such as facilities that meet a condition input by a user.

  Conventionally, as a device for searching for facilities (point candidates) based on a plurality of keywords (character strings) input by a user, for example, a point search device described in Patent Document 1 has been proposed. In the point search device described in Patent Document 1, when a plurality of keywords are input, an AND search is performed using the first keyword and the second keyword, and the facility matches the conditions indicated by the keywords. Is displayed as a search result.

  Such an AND search is performed when each keyword is input according to a predetermined rule set in advance. For example, the first keyword input first is a character string indicating a facility to be searched or a part thereof, while the second keyword input later is a region where the facility to be searched exists. The character string to be shown or a part thereof. That is, when “Tokyo” is input as the first keyword and “Uraya” is input as the second keyword, an area having a name that matches “Uraya” (for example, “Urayasu City”) is predicted. . Then, in each facility existing in the predicted area, a facility (for example, “Tokyo Disneyland”) that directly matches “Tokyo” is searched, and the search result is displayed on the display.

JP 2009-266171 A

  By the way, in the point search apparatus described in Patent Document 1, when a plurality of keywords are input according to the predetermined rule, it is supposed that the time required for input and search can be shortened. On the other hand, when inputting a plurality of keywords ignoring predetermined rules, the effect of shortening the time required for input and search cannot be expected.

  The present invention has been made in view of such circumstances. The purpose is to provide a point search device, a point search method, and a point search program that can reduce the time required for the user to input a character string when inputting a plurality of character strings.

In order to achieve the above object, the present invention provides a point search device that searches for point candidates that match the conditions indicated by a plurality of input character strings, and includes a plurality of registered character strings and each of the registered character strings. Storage means for storing attribute information indicating an attribute, and when a registered character string corresponding to a confirmed character string is stored in the storage means, the attribute information of the registered character string is assigned to the attribute of the confirmed character string. Read from the storage means an attribute acquisition means for acquiring information, a registered character string that at least partially matches a character string before confirmation that is input subsequent to the confirmed character string, and attribute information of the registered character string. And a registration means that does not have attribute information that matches the attribute information acquired by the attribute acquisition means among the registered character strings read by the read means when performing an AND search with a plurality of character strings. Sentence Prediction display means for displaying a column on the display screen as a prediction candidate of the character string before confirmation in preference to a registered character string having matching attribute information, and the prediction displayed on the display screen by the prediction display means A replacement display unit that, when a candidate is selected, replaces the character string before confirmation with the prediction candidate and displays it on the display screen, the storage unit includes the plurality of registered character strings, and the registration The attribute information for each character string and the prediction priority set for each registered character string are stored in association with each other, and the prediction display unit performs the AND search with a plurality of character strings. A correction process for correcting the priority of the registered character string is performed based on the attribute information acquired by the attribute acquiring unit, and the correction process is performed among the registered character strings read from the storage unit by the reading unit. High registered character string priority after, is summarized in that to be displayed on the display screen as the predicted candidates of the correction process after the lower priority of registered character before the deterministic preferentially than Strings.

  According to the above configuration, when input of the next character string is started when there is a confirmed character string, the character string being input, that is, the prediction candidate of the character string before confirmation is displayed on the display screen. . In order to realize such processing, when a character string corresponding to a confirmed character string is stored as a registered character string in the storage means, the attribute information of the registered character string is attribute information ( Hereinafter, it is also read as “determined attribute information”. Then, when the input of the next character string is started, the registered character string that at least partially matches the character string before the determination and the attribute information of the registered character string are read from the storage unit. Subsequently, at least one registered character string among the read registered character strings is displayed on the display screen as a prediction candidate of the character string before being confirmed. That is, when performing a search, a prediction candidate for a character string before confirmation is set based on the attribute information of the character string that has already been confirmed.

  In particular, when performing an AND search with a plurality of character strings, among the registered character strings read from the storage means, the registered character string having no definite attribute information has priority over the registered character string having definite attribute information. Are displayed on the display screen as prediction candidates. This is because when performing an AND search, it is considered unlikely that a character string having the same attribute information as the confirmed attribute information of the confirmed character string is input. And the prediction candidate selected by the user among the prediction candidates displayed on the display screen is replaced with the character string before finalization. Therefore, it is possible to reduce the time required for the user to input a character string when inputting a plurality of character strings.

Moreover, according to the said structure, the priority of the registration character string which has the said definite attribute information is correct | amended. Then, the registered character string having a high priority after the correction process is displayed on the display screen as a prediction candidate of the character string before being determined in preference to the registered character string having a low priority after the correction process. Therefore, the prediction accuracy of the character string in the middle of input can be improved, and the possibility of contributing to shortening the time required for inputting the character string is increased.
In point searching apparatus of the present invention, before Symbol predictive display means, when performing an AND search multiple strings, the priority of the registered character string having the attribute information matching the attribute information acquired by the attribute acquisition unit the intends line the correction processing to lower.

According to the above configuration, the priority of the registered character string having the definite attribute information is corrected to be low. Then, the registered character string having a high priority after the correction process is displayed on the display screen as a prediction candidate of the character string before being determined in preference to the registered character string having a low priority after the correction process. Therefore, the prediction accuracy of the character string in the middle of input can be improved, and the possibility of contributing to shortening the time required for inputting the character string is increased.
In the point search apparatus according to the present invention, the prediction display means determines the priority of the registered character string having attribute information that does not match the attribute information acquired by the attribute acquisition means when performing AND search with a plurality of character strings. Perform correction processing to increase.

  In the point search device of the present invention, when the AND display is performed with a plurality of character strings, the prediction display means displays attribute information of a higher concept than the attribute information of the confirmed character string acquired by the attribute acquisition means. A registered character string that does not exist is displayed on the display screen as a prediction candidate of the character string before the determination, with priority over the registered character string having the attribute information of the superordinate concept.

  In the case of performing an AND search, it is considered that a user is unlikely to input a character string having higher-level attribute information than the above-described fixed attribute information of a fixed character string. Therefore, in the present invention, a registered character string having higher concept attribute information than the above-mentioned fixed attribute information is displayed on the display screen as a prediction candidate of a character string before finalization than a registered character string having no higher concept attribute information. It becomes difficult to display. For this reason, it is possible to improve the prediction accuracy of a character string in the middle of input as much as the possibility that a character string that is unlikely to be input by the user is displayed as a prediction candidate on the display screen can be reduced. Therefore, the possibility that the time required for inputting the character string can be shortened is increased.

  In the point search device of the present invention, the prediction display means, when performing AND search with a plurality of character strings, the attribute information of the determined character string is attribute information related to the area name, and the determined character When there is only one candidate for the area indicated by the column, a registered character string having no attribute information related to the area name in a wider area than the area is used as a registered character string having attribute information related to the area name in the wide area. The character string is displayed on the display screen as a prediction candidate of the character string before being determined in preference to the character string.

  When there is only one candidate for the region indicated by the confirmed character string, there is a low possibility that a character string indicating a region indicating a wider area than the region is input. Therefore, in the present invention, when there is only one candidate for the area indicated by the confirmed character string, the registered character string that does not have the attribute information regarding the area name of a wider area than the area is the above wide area. Is displayed as a prediction candidate preferentially over a registered character string having attribute information regarding the area name. Therefore, the prediction accuracy of the character string in the middle of input can be improved as much as the possibility that a character string that is unlikely to be input by the user can be displayed as a prediction candidate can be reduced. Therefore, the possibility that the time required for inputting the character string can be shortened is increased.

  In the point search apparatus according to the present invention, the prediction display means, when performing a NOT search with a plurality of character strings, from the attribute information acquired by the attribute acquisition means among the registered character strings read by the reading means. In addition, a registered character string having lower-level concept attribute information is displayed on the display screen as a prediction candidate of the character string before determination, with priority over a registered character string not having the lower-level concept attribute information.

  When performing a NOT search with a plurality of character strings, among the registered character strings read from the storage means, the registered character string having the attribute information of the subordinate concept of the definite attribute information has the attribute information of the subordinate concept. It is displayed on the display screen with priority over no registered character string. This is because, when a NOT search is performed, there is a high possibility that the attribute information of the character string before the determination is attribute information of a subordinate concept of the determination attribute information due to the characteristics of the NOT search. In other words, in the present invention, even when a NOT search is performed, a prediction candidate for a character string before confirmation is set based on the attribute information of the character string that has already been confirmed. Therefore, it is possible to reduce the time required for the user to input a character string when inputting a plurality of character strings.

  In the point search apparatus of the present invention, the storage unit stores the plurality of registered character strings, the attribute information for each registered character string, and the prediction priority set for each registered character string in association with each other. The predictive display means, when performing a NOT search with a plurality of character strings, has a registered character string having attribute information of a lower concept than the attribute information of the confirmed character string acquired by the attribute acquisition means. Of the registered character strings read out from the storage means by the reading means, the registered character strings having a high priority after the correction processing are set to the priority values after the correction processing. It is displayed on the display screen as a prediction candidate of the character string before the determination in preference to a lower registered character string.

  According to the above configuration, when a NOT search is performed, the priority of a registered character string having attribute information of a lower concept than the determined attribute information is corrected to be higher. Then, the registered character string having a high priority after the correction process is displayed on the display screen as a prediction candidate of the character string before being determined in preference to the registered character string having a low priority after the correction process. Therefore, even when a NOT search is performed, the accuracy of predicting a character string in the middle of input can be improved, and the possibility of contributing to shortening the time required to input a character string is increased.

The present invention is a point search method by a point search device provided with a control unit that searches for point candidates that match the conditions indicated by a plurality of input character strings, wherein the control unit includes a plurality of registered character strings, When a registered character string corresponding to a confirmed character string is stored in the storage means for storing attribute information indicating the attribute of each registered character string, the attribute information of the registered character string is set to the confirmed character string. An attribute acquisition step for acquiring from the storage means as attribute information, a registered character string that at least partially matches a character string before confirmation that is input subsequent to the confirmed character string, and attribute information of the registered character string When performing an AND search using a plurality of character strings and a reading step to be read from the storage means, the attribute information acquired in the attribute acquisition step is the same as the registered character string read in the reading step. A prediction display step of displaying a registered character string having no sex information on a display screen as a prediction candidate of the character string before the determination with priority over a registered character string having matching attribute information; and the prediction display step wherein when the display user predictive candidates displayed on the screen is selected, a character string before confirmation have a, and replacement display step of displaying on the display screen is replaced on the estimated candidates, the storage means, the A plurality of registered character strings, attribute information for each registered character string, and a prediction priority set for each registered character string are stored in association with each other. In the prediction display step, a plurality of character strings are stored. When performing an AND search, a correction process for correcting the priority of the registered character string is performed based on the attribute information acquired in the attribute acquisition step, and the reading means reads from the storage means. Among the registered character strings, the registered character string having a high priority after the correction process is preferentially used as a prediction candidate of the character string before the determination, in preference to the registered character string having a low priority after the correction process. The gist is to display on the display screen .

According to the said structure, the effect | action and effect equivalent to the said point search apparatus can be acquired.
The present invention is a point search program for causing a control device to search for point candidates that match the conditions indicated by a plurality of input character strings, the control device including a plurality of registered character strings , Storage means for storing attribute information indicating an attribute of a registered character string and a prediction priority set for each registered character string in association with each other, and a registered character string corresponding to a confirmed character string in the storage means When stored, attribute acquisition means for acquiring the attribute information of the registered character string as attribute information of the confirmed character string, and at least a character string before confirmation that is input subsequent to the confirmed character string A registered character string that partially matches and a reading unit that reads out attribute information of the registered character string from the storage unit, and when performing an AND search on a plurality of character strings, the registered character string read by the reading unit Before A registered character string that does not have attribute information that matches the attribute information acquired by the attribute acquisition means is preferentially displayed on the display screen as a prediction candidate of the character string before the determination over the registered character string that has matching attribute information. A prediction display unit to be displayed; and a replacement display unit that, when a prediction candidate displayed on the display screen by the prediction display unit is selected, replaces the character string before the confirmation with the prediction candidate and displays it on the display screen. Correction processing for correcting the priority of the registered character string based on the attribute information acquired by the attribute acquisition means when the control device performs an AND search with a plurality of character strings as the prediction display means. Among the registered character strings read from the storage unit by the reading unit, the registered character string having a high priority after the correction process is added to the supplementary character string. And summarized in that to be displayed on the display screen as the predicted candidate character string before confirmation preference to lower registered character string priority after treatment.

  According to the said structure, the effect | action and effect equivalent to the said point search apparatus can be acquired.

The block diagram explaining one Embodiment of the navigation apparatus of a vehicle provided with the point search apparatus of this invention. Explanatory drawing which shows an example of the display mode of a display at the time of searching a point candidate. Explanatory drawing which shows an example of the display mode of a display at the time of searching a point candidate by AND search. Explanatory drawing which shows an example of the display mode of a display at the time of searching a point candidate by AND search. Explanatory drawing explaining the one aspect | mode of a prediction word table. The flowchart explaining a definite attribute determination processing routine. The flowchart explaining the prediction candidate determination processing routine for AND search. The flowchart explaining a replacement process routine. Explanatory drawing explaining the one aspect | mode of a prediction word table. The flowchart explaining a definite attribute determination processing routine. The flowchart explaining the prediction candidate determination processing routine for AND search. Explanatory drawing explaining the one aspect | mode of a prediction word table. The flowchart explaining a definite attribute determination processing routine. The flowchart explaining the prediction candidate determination processing routine for AND search. Explanatory drawing which shows an example of the display mode of a display at the time of searching a point candidate by NOT search. The flowchart explaining the prediction candidate determination process routine for NOT search. Explanatory drawing explaining the one aspect | mode of a prediction word table.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described with reference to FIGS.
As shown in FIG. 1, the vehicle navigation device 11 includes a display device 12 and a control device 20 as an example of a point search device that controls the display device 12. The display device 12 includes a display 13 as an example of a display screen that provides various information to a vehicle occupant (user) in a visually recognizable state, and a liquid crystal touch panel (input means) 14 disposed in front of the display 13. And are provided. Further, a plurality of types of operation buttons (input means) 15 are provided on the side of the display 13 in the display device 12.

  When such an operation of the operation button 15 and the touch panel 14 by the occupant is accepted, the display device 12 transmits a signal corresponding to the operation content to the control device 20. The display device 12 receives the image data transmitted from the control device 20 as a response to the transmitted signal, and causes the display 13 to display an image based on the image data.

  The position detection unit 30 is connected to the control device 20 via the vehicle-side I / F 21 in a state where information can be transmitted and received. The position detector 30 includes a GPS (Global Positioning System) receiver 31, a vehicle speed sensor 32, and a gyro sensor 33. The control device 20 detects an absolute position such as latitude / longitude by radio navigation based on the received signal received by the GPS receiver 31. Further, the control device 20 calculates a relative position from the reference position using autonomous navigation based on detection signals from the vehicle speed sensor 32 and the gyro sensor 33. And the control apparatus 20 specifies the present location of the own vehicle combining the absolute position and relative position which are output from the position detection part 30. FIG.

  The control device 20 includes a control unit 22, an image processing processor 23, an information storage device 24, a table storage unit 25 as an example of a storage unit, and the like. The control unit 22 is provided with a digital computer having a CPU 22a, a ROM 22b, a RAM 22c, and the like. The ROM 22b has a program (point search program) for searching for point candidates (for example, beaches and restaurants) that match the facility attributes selected by the occupant, and a program for searching and guiding a route to the destination. Etc. are stored. In the RAM 22c, a storage area for storing a work area of the CPU 22a, a searched route, and the like is formed.

  The image processor 23 generates various image data to be displayed on the display 13 based on the control signal from the control unit 22, and transmits the generated image data to the display device 12. For example, when the control unit 22 searches for a route from the current position of the host vehicle to the destination, the image processing processor 23 displays a searched route and a screen for requesting permission to start guidance along the route. Is generated, and an image based on the image data is displayed on the display 13. When the target point candidate is searched by the control unit 22, the image processor 23 generates image data for guiding the searched point candidate and displays an image based on the image data on the display 13. Let

  The information storage device 24 is an external storage medium such as a hard disk or an optical disk. In the information storage device 24, a map data storage unit 24a, a search data storage unit 24b, and a facility data storage unit 24c are formed. The map data storage unit 24a stores map data such as road maps, house maps, and building shape maps. The search data storage unit 24b stores data relating to various road information such as a road between predetermined points and traffic regulation information of the road, distance, and toll. The facility data storage unit 24c stores various facilities such as a ward office, a hospital, a station, a restaurant, an amusement park, and a parking lot, and data such as addresses and telephone numbers of the various facilities.

  The table storage unit 25 is constructed by a nonvolatile memory or the like. The table storage unit 25 stores a predicted word table shown in FIG. 5 that is used when searching for point candidates.

Next, an example of the search display screen displayed on the display 13 when searching for a point candidate will be described with reference to FIGS. 2, 3, and 4. FIG.
As shown in FIG. 2, the display 13 displays an inputable character area 40 in which characters selected by a vehicle occupant are displayed, and a character string (word) composed of characters selected by the occupant. A selected character display area 41 is formed. When an area in which “A” is written in the input character area 40 is selected, “A” is displayed in the selected character display area 41. Further, on the right side of the selected character display area 41 on the display 13, a correction area 42 that is selected when deleting the selected characters one by one, a search execution area 43 that is selected when executing the search, An AND region 44 that is selected when an AND search is performed using a plurality of words and a NOT region 45 that is selected when a NOT search is performed using a plurality of words are sequentially arranged.

  For example, when performing an AND search with a plurality of words, the AND region 44 is selected after the input of the first word (in this case, “Ah”) is completed. Then, as shown in FIG. 3, the logical character “&” for AND search is displayed in the selected character display area 41 following the word “that”. Then, the first word is determined as “Ah”. That is, a character string (in this case, “Ah”) displayed before the logical operator “&” (left side in FIG. 3) corresponds to a confirmed character string.

  When the input of the second word is started following the logical operator “&”, the character displayed on the right side of the logical operator “&” is displayed on the right side of the input-capable character area 40 on the display 13. A predicted word display area 46 in which prediction candidates estimated from the columns are displayed is formed. In this way, the character string (“A” in FIG. 3) displayed on the right side of the logical operator “&” corresponds to the character string before confirmation. In the predicted word display area 46, as will be described in detail later, among the registered words stored in the predicted word table shown in FIG. For example, when the character string before confirmation is “A”, the predicted word display area 46 displays a word that matches “A” in front as a prediction candidate. Here, when a character “U”, for example, is selected after “A”, the predicted word display area 46 displays a word that matches the character string “AU” that is the character string before confirmation as a prediction candidate. The In addition, when the word that coincides with the character string before confirmation does not exist in the predicted word table, the predicted word display area 46 is not formed on the display 13.

  Then, as shown in FIG. 4, when any one of a plurality of prediction candidates (for example, “Outlet”) displayed in the predicted word display area 46 is selected, the selected character display area 41 displays Is replaced with the selected prediction candidate (outlet). As a result, “Any & Outlet” is displayed in the selected character display area 41. When the search execution area 43 is selected in this state, an AND search is performed on the first word (that) and the second word (outlet). The “AND search” is to extract a search result including all of a plurality of input words.

  As described above, when searching for a point candidate using the navigation device 11 of the present embodiment, a prediction candidate for a character string before being confirmed is guided and displayed on the display 13. Therefore, next, a prediction word table used when guiding and displaying such prediction candidates on the display 13 will be described with reference to FIG.

  As shown in FIG. 5, the predicted word table includes a plurality of registered words (registered character strings) arranged in accordance with a predetermined rule (in this embodiment, in Japanese alphabetical order) and attributes (attributes) for each registered word. Information) and the score value S for each registered word are stored in association with each other. Note that the score value S indicates the priority of prediction, and a registered word with a high score value S is more likely to be displayed on the display 13 as a prediction candidate compared to a registered word with a low score value S.

Next, a part of each registered word stored in the predicted word table will be described in detail.
Since “AA” is a company name, the attribute “AA” is a facility name, and the score value S of “AA” is “260”. Since “ice” is an abbreviation for ice cream, the attribute of “ice” is a genre name (type of food), and the score value S of “ice” is “380”.

  “Ayu” is used to indicate the area “A prefecture”, and is also used to indicate a facility name such as “Aoi Park”. Therefore, “Ayu” has an attribute called facility name and an attribute called address name (prefecture). The score value “S” for such “that” is “580”. “Outlet” is an abbreviation for “Outlet Mall”. In the outlet mall, fashion-related stores and food-related stores open. Therefore, “Outlet” has an attribute of a genre name (fashion relation / meal relation), and the score value S of “Outlet” is “380”.

  “Iai” is used to represent the area “Iai Prefecture”, and is also used to represent the name of a facility such as “Iai Park”. Therefore, “Iai” has an attribute called facility name and an attribute called address name (prefecture). The score value S of such “Iai” is “450”. “Ooiyu” is used to represent an area called “Ooisai City” and also used to represent a facility name such as “Ooisai Building”. For this reason, “Oh!” Has an attribute called facility name and an attribute called address name (city). The score value S of such “oh” is “400”.

  Since “sushi (sushi)” is a word representing food, the attribute of “sushi” is a genre name (type of food), and the score value S of “sushi” is “350”. “Tatasushi” is the name of a store where you can eat sushi. Therefore, “Tatashi” has an attribute called a facility name and an attribute called a genre name (food shop). The score value “S” for such “tasushi” is “200”.

  Although not shown in FIG. 5, verbs such as “eat (eat)” and “play (play)” are also stored as registered words in the predicted word table. As attributes for such verbs, attributes that match the meaning of the verbs are set. For example, the attributes “eat (eat)” and “play (play)” are set in the genre name.

  Next, a processing routine executed by the control unit 22 when the vehicle occupant searches for point candidates will be described based on the flowcharts shown in FIGS. 6, 7, and 8. FIG. 6 is a flowchart for explaining a fixed attribute determination processing routine for acquiring the attribute of the word indicated by the fixed character string as a fixed attribute. FIG. 7 illustrates a prediction candidate determination processing routine for guiding and displaying on the display 13 a word input after the input of the logical operator “&” for AND search, that is, a prediction candidate of a character string before being confirmed. It is a flowchart. FIG. 8 is a flowchart for explaining a replacement processing routine that is executed when a prediction candidate is guided and displayed on the display 13.

First, the fixed attribute determination processing routine will be described with reference to the flowchart shown in FIG.
This fixed attribute determination processing routine is executed at a timing when the AND area 44 or the NOT area 45 (see FIG. 2) is selected by the vehicle occupant. Then, in the confirmed attribute determination processing routine, the control unit 22 determines whether or not a word corresponding to the confirmed character string is in the predicted word table (step S110). As an example, the control unit 22 searches the predicted word table for a registered word that matches the confirmed character string in the forward direction. Note that the search at this time may be an ambiguous search in which objects whose search conditions do not completely match are extracted according to a certain rule. Then, the searched registered word is set as “a word corresponding to a confirmed character string”.

  When the word corresponding to the confirmed character string is in the predicted word table (step S110: YES), the control unit 22 reads the attribute of the registered word corresponding to the confirmed character string from the predicted word table, and stores the attribute. Obtained as a definite attribute (step S111). For example, when the word corresponding to the confirmed character string is “sushi”, the control unit 22 reads “genre name (food type)” as the attribute of the registered word “sushi”, and “genre name (food type)”. Type) ”is acquired as a definite attribute (see FIG. 5). When the word corresponding to the confirmed character string is “Ayu”, the control unit 22 reads “facility name” and “address name (prefecture)” as the attribute of the registered word “Aoi”. "Name" and "Address name (prefecture)" are acquired as fixed attributes. Therefore, in the present embodiment, the control unit 22 also functions as an attribute acquisition unit. Step S111 corresponds to an attribute acquisition step. Then, the control part 22 complete | finishes a definite attribute determination processing routine.

  On the other hand, when the word corresponding to the confirmed character string is not in the predicted word table (step S110: NO), the control unit 22 sets the facility name as the confirmed attribute (step S112), and then the confirmed attribute determination processing routine. Exit. This is because a word having the facility name as an attribute is more likely not to be covered in the predicted word table than a word having the genre name or address name as an attribute. In particular, a word indicating the name of a new building or the like is very likely not stored in the predicted word table.

Next, the prediction candidate determination processing routine will be described with reference to the flowchart shown in FIG.
This prediction candidate determination processing routine is executed each time a character is selected by a vehicle occupant after the input of the logical operator “&” for AND search. Further, the prediction candidate determination processing routine is executed when a character string before confirmation exists even at the timing when the correction area 42 is operated by the occupant. Then, in the prediction candidate determination processing routine, the control unit 22 includes, in each registered word included in the predicted word table, a registered word that matches the character string before confirmation that is input following the confirmed character string. It is determined whether or not there is (step S200). That is, the control unit 22 determines whether or not there is a character string before confirmation among the input character strings, and if there is a character string before confirmation, the registered word that matches the confirmed character string forward It is determined whether or not there is. If there is no registered word that matches forward with the character string before confirmation (step S200: NO), the control unit 22 determines that prediction using the prediction word table cannot be performed, and ends the prediction candidate determination processing routine. . At this time, the control unit 22 also ends the prediction candidate determination processing routine even when there is no character string before confirmation among the input character strings. In addition, when the prediction candidate is displayed on the display 13 before the execution of the current prediction candidate determination processing routine is started, the control unit 22 deletes the predicted word display area 46 (see FIG. 3) from the display 13 and then Then, the prediction candidate determination processing routine is terminated.

  On the other hand, when there is a registered word that matches forward with the character string before confirmation (step S200: YES), the control unit 22 determines the registered word that matches forward with the character string before confirmation and the attribute of the registered word from the predicted word table. All are read (step S201). For example, when the character string before confirmation is “A”, the control unit 22 predicts “AO”, “ICE”, “ANY” and “OUTLET” together with their attributes as a registered word that matches “A” in front. Read from the word table (see FIG. 5). When the character string before confirmation is “Ai”, the control unit 22 reads “Ice” and “Ah” from the predicted word table together with their attributes as registered words that match the front of “Ai” ( (See FIG. 5). Therefore, in this embodiment, the control unit 22 also functions as a reading unit. Step S201 corresponds to a reading step.

  Subsequently, the control unit 22 acquires the total number of registered words read from the predicted word table as a read total number P (step S202), and sets “1” to the coefficient M (step S203). Then, the control unit 22 arranges the registered words read in step S201 in alphabetical order, and acquires the score value S (M) of the Mth registered word from the top from the predicted word table (step S204). Subsequently, the control unit 22 determines whether or not the attribute of the Mth registered word matches the confirmed attribute set in step S111 or step S112 (step S205). At this time, when a plurality of attributes (for example, a facility name and an address name (prefecture)) are set as the confirmed attributes, the control unit 22 determines that the attribute of the Mth registered word is any one of the confirmed attributes. It is determined whether or not they match. In the present embodiment, it is assumed that “ice” and “outlet” have the same attribute as a genre name. In addition, “Ayu”, “Iai” and “Ooi” have the same attribute of address name.

  When there is no fixed attribute that matches the attribute of the Mth registered word (step S205: NO), the control unit 22 uses the score value S (M) of the Mth registered word as the corrected score value of the Mth registered word. SH (M) is set (step S206), and the process proceeds to step S208 described later. On the other hand, when there is a definite attribute that matches the attribute of the Mth registered word (step S205: YES), the control unit 22 performs a correction process to lower the score value S (M) of the Mth registered word (step S205). S207). In the present embodiment, the control unit 22 subtracts the score value S (M) of the Mth registered word by a preset correction value H (for example, 50), and the subtraction result is corrected score value SH (M). And Thereafter, the control unit 22 moves the process to the next step S208. The correction value H may be a value based on the score value S of a word corresponding to a confirmed character string.

  In step S208, the control unit 22 determines whether or not the current coefficient M matches the total number P read out acquired in step S202. When the coefficient M is less than the total number P to be read (step S208: NO), the control unit 22 has not performed the determination process of step S205 for all registered words read from the predicted word table, so the coefficient M is set to “1”. (Step S209), and the process proceeds to step S204 described above. On the other hand, when the coefficient M matches the read total number P (step S208: YES), the control unit 22 determines that the determination process of step S205 has been performed for all registered words read from the predicted word table. Thus, when the determination result of step S208 becomes affirmative, among each registered word read from the predicted word table, the score value of the registered word having an attribute that matches the confirmed attribute (in this case, the corrected score value SH) ) And the score value (in this case, the corrected score value SH) of other registered words that do not have an attribute that matches the confirmed attribute.

  Then, the control unit 22 arranges the registered words read from the predicted word table in descending order of the correction score value SH, and extracts the top N registered words (N is an integer of 1 or more, for example, “5”). (Step S210). Subsequently, the control unit 22 performs prediction candidate display processing for guiding and displaying the top N registered words extracted in step S210 on the display 13 as prediction candidates (step S214), and then ends the prediction candidate determination processing routine. To do. In this case, a registered word having a high correction score value SH corresponding to the priority after the correction process is preferentially displayed on the display 13 as a prediction candidate over a registered word having a low correction score value SH. Therefore, in this embodiment, the control part 22 functions also as a prediction display means. Moreover, a prediction display step is comprised by step S210, S214.

Next, the replacement processing routine will be described with reference to the flowchart shown in FIG.
The replacement processing routine is executed at predetermined intervals (for example, every 0.1 second) while the prediction candidates are displayed on the display 13. In the replacement processing routine, the control unit 22 determines whether or not a prediction candidate displayed in the predicted word display area 46 (see FIG. 3) formed on the display 13 has been selected by the vehicle occupant (step S300). ). When the prediction candidate is not selected (step S300: NO), the control unit 22 once ends the replacement processing routine.

  On the other hand, when a prediction candidate is selected (step S300: NO), the control unit 22 specifies the selected prediction candidate and performs a replacement process (step S301). Specifically, the control unit 22 replaces a character string (for example, a) before confirmation displayed in the selected character display area 41 of the display 13 with the selected prediction candidate (for example, an outlet) (FIG. 3). , 4). Therefore, in the present embodiment, the control unit 22 also functions as a replacement display unit. Step S301 corresponds to a replacement display step. Then, the control unit 22 deletes the predicted word display area 46 from the display 13 (step S302), and then ends the replacement processing routine once.

Therefore, in this embodiment, the following effects can be obtained.
(1) In this embodiment, when a word is input following a logical operator, a character string in the middle of input, that is, a prediction candidate of a character string before confirmation is displayed on the display 13. When the prediction candidate displayed as a guide is selected by the vehicle occupant, the character string before confirmation is replaced with the prediction candidate selected by the occupant. Then, when the search execution area 43 of the display 13 is selected in this state, a search using a plurality of words is started. Therefore, when an occupant selects a word guided and displayed on the display 13 as a prediction candidate, the time required for input can be shortened.

  (2) In this embodiment, when a word corresponding to a confirmed character string is stored as a registered word in the predicted word table, the attribute of the registered word is acquired as a confirmed attribute. Then, when the input of the next word is started with the logical operator “&” selected to perform an AND search, the registered word that matches the character string before confirmation and the attribute of the registered word are predicted. Read from the word table. Then, among the read registered words, a registered word that does not have an attribute that matches the confirmed attribute over a registered word that has an attribute that matches the confirmed attribute is preferentially displayed as a prediction candidate for the character string before confirmed. Is displayed on the screen. This is because, when performing an AND search, it is considered that there is a low possibility that words having the same attribute are continuously input. That is, prediction candidates that are highly likely to be selected by a vehicle occupant are displayed on the display 13 with high accuracy. Therefore, it is possible to shorten the time required for inputting words by the vehicle occupant when inputting a plurality of words.

  (3) In the present embodiment, among the plurality of registered words read from the predicted word table, the priority (that is, the corrected score value SH) of the registered character string having an attribute that matches the confirmed attribute is lowered. On the other hand, the priority (that is, the corrected score value SH) of the registered character string that does not have an attribute that matches the confirmed attribute is maintained at the priority (that is, the score value S) before correction. Then, a registered word with a high correction score value SH is preferentially displayed on the display 13 as a prediction candidate for a character string before being confirmed, with priority over a registered word with a low correction score value SH. Therefore, the prediction accuracy for the character string before confirmation can be improved.

  (4) In the present embodiment, when a plurality of registered words are read from the predicted word table based on the character string before confirmation, the correction of the score value S based on whether or not the attribute matches the confirmed attribute. Processing is performed. Thereafter, among the registered words read out as prediction candidates, the top N registered words having the highest correction score value SH are displayed on the display 13 as prediction candidates. At this time, when the number of registered words read out as prediction candidates is larger than N, registered words having a low correction score value SH are not guided and displayed on the display 13 as prediction candidates. Therefore, it is possible to reduce the possibility that a registered word that is unlikely to be selected by a vehicle occupant, that is, a registered word having an attribute that matches the confirmed attribute, is displayed on the display 13 as a prediction candidate.

  (5) If the word corresponding to the character string already confirmed at the stage when the logical operator “&” is input is not in the predicted word table, the confirmed attribute is the facility name. As a result, the confirmed character string and the character string before confirmation that is input following the logical operator “&” can be predicted using the predicted word table.

(Second Embodiment)
Next, a second embodiment of the present invention will be described with reference to FIGS. The second embodiment is different from the first embodiment in the contents of the predicted word table, the definite attribute determination processing routine, and the prediction candidate determination processing routine. Therefore, in the following description, parts different from those of the first embodiment will be mainly described, and the same or corresponding member configurations as those of the first embodiment are denoted by the same reference numerals, and redundant description will be omitted. Shall.

First, the predicted word table of this embodiment will be described with reference to FIG.
As shown in FIG. 9, in the predicted word table, score values S1, S2, and S3 are set for each attribute for each registered word. Specifically, the registered word “Ah” representing a company name has an attribute of facility name. Therefore, the first score value S1 for the facility name for “Ah” is set to “260”, while the second score value S2 for the address name and the third score value S3 for the genre name are both “0 (zero)”. " In addition, “ice”, which is an abbreviation for ice cream, has an attribute of genre name (type of food). Therefore, the third score value S3 for “ice” is “380”, while the first score value S1 and the second score value S2 are both “0 (zero)”.

  In addition, as described above, “Ayu” has an attribute called facility name and an attribute called address name (prefecture). For this reason, the first score value S1 corresponding to “Yes” is set to “348”, the second score value S2 is set to “232”, and the third score value S3 is set to “0 (zero)”. Further, “Outlet”, which is an abbreviation for Outlet Mall, has an attribute of genre name (fashion relation / meal relation). Therefore, the third score value S3 for “Outlet” is set to “380”, while the first score value S1 and the second score value S2 are both set to “0 (zero)”.

  Further, as described above, “Iai” has an attribute called facility name and an attribute called address name (prefecture). Therefore, the first score value S1 and the second score value S2 for “Iai” are both “225”, while the third score value S3 is “0 (zero)”. In addition, as described above, “oh” has an attribute called facility name and an attribute called address name (city). Therefore, the first score value S1 for “Oh” is set to “160”, the second score value S2 is set to “240”, and the third score value S3 is set to “0 (zero)”. .

  Further, “sushi (sushi)” has an attribute of genre name (type of food). Therefore, the third score value S3 for “sushi” is “350”, while the first score value S1 and the second score value S2 are both “0 (zero)”. Further, “Tatasushi” representing the name of a store where sushi can be eaten has an attribute called a facility name and an attribute called a genre name (food shop). Therefore, the first score value S1 and the third score value S3 for “addition” are both “100”, while the second score value S2 is “0 (zero)”.

Next, the fixed attribute determination processing routine of this embodiment will be described with reference to the flowchart shown in FIG.
In the confirmed attribute determination processing routine, when a word corresponding to the confirmed character string is in the predicted word table (step S110: YES), the control unit 22 determines the first score value S1 and the first score of the registered word searched. 2 score value S2 and 3rd score value S3 are acquired from a prediction word table (step S121). For example, when the word corresponding to the confirmed character string is “Oh,” the control unit 22 sets the first score value S1 (160), the second score value S2 (240), and the third of “Oh!” A score value S3 (0 (zero)) is acquired. In the present embodiment, a score value is set for each attribute. Therefore, acquiring each score value S1, S2, S3 of the registered word that matches the condition is equivalent to acquiring the attribute (attribute information) of the registered word. Therefore, in the present embodiment, step S121 corresponds to an attribute acquisition step.

  Subsequently, the control unit 22 sets the acquired first score value S1 as the first correction value H1 for the facility name, sets the second score value S2 as the second correction value H2 for the address name, and The third score value S3 is set to the third correction value H3 for genre name (step S122). Then, the control part 22 complete | finishes a definite attribute determination processing routine.

  On the other hand, when the word corresponding to the confirmed character string is not in the predicted word table (step S110: NO), the control unit 22 sets the first correction value H1, the second correction value H2, and the third correction value H3 to “0”. (Zero) "is set (step S123), and then the definite attribute determination processing routine ends.

Next, the prediction candidate determination processing routine of this embodiment will be described with reference to the flowchart shown in FIG.
In the prediction candidate determination process routine, the control unit 22 sequentially executes processes equivalent to the processes in steps S200, S201, S202, and S203. Subsequently, the control unit 22 arranges the registered words read out in step S201 in the order of the Japanese syllabary, and predicts the score values S1 (M), S2 (M), and S3 (M) of the Mth registered word from the top. Obtained from the word table (step S221). The control unit 22 then corrects the correction values H1, H2, and H3 set in step S122 or S123 for each score value S1 (M), S2 (M), and S3 (M) of the acquired Mth registered word. A correction process using is performed (step S222).

  Specifically, the control unit 22 subtracts the first correction value H1 from the first score value S1 (M) of the Mth registered word, and sets the subtraction result as the first correction score value SH1 (M). Further, the control unit 22 subtracts the second correction value H2 from the second score value S2 (M), and sets the subtraction result as the second correction score value SH2 (M). Then, the control unit 22 subtracts the third correction value H3 from the third score value S3 (M), and sets the subtraction result as the third correction score value SH3 (M). If the corrected score value SH becomes a negative value by the above calculation, the corrected score value SH is set to “0 (zero)”.

  Subsequently, the control unit 22 adds the correction score values SH1 (M), SH2 (M), and SH3 (M) calculated in Step S222, and sets the sum as a total score value SHT (M) (Step S22). S223).

  Here, a case where the word corresponding to the confirmed character string is “Ah” and the Mth registered word is “Tatoshi” will be described as an example. That is, in step S122, the first correction value H1 is set to “348”, the second correction value H2 is set to “232”, and the third correction value H3 is set to “0 (zero)”. (See FIG. 9). Further, the first score value S1 (M) and the third score value S3 are set to “100”, and the second score value S2 is set to “0 (zero)”. Then, as a result of the correction process in step S222, the first correction score value SH1 (M) is “0 (zero)” because the calculation result ((100-348) <0 (zero)) becomes a negative value. Become. The second corrected score value SH2 (M) is “0 (zero)” because the calculation result ((0 (zero) −232) <0 (zero)) is a negative value. In addition, the third corrected score value SH3 (M) is “100” (= 100−0 (zero)). The total score value SHT (M) is “100” (= 0 (zero) +0 (zero) +100).

  Then, similarly to the determination process in step S208, the control unit 22 determines whether or not the current coefficient M matches the total read number P acquired in step S202 (step S224). When the coefficient M is less than the total number P to be read (step S224: NO), the control unit 22 increments the coefficient M by “1” (step S225), and the process proceeds to step S221 described above. On the other hand, when the coefficient M coincides with the total number P to be read (step S224: YES), the control unit 22 sequentially executes the processes of steps S210 and S214, and ends the prediction candidate determination process routine.

  Here, as in the case of the first embodiment, when the score value of the registered word is not set for each attribute, the following problem may occur. As a premise, it is assumed that an AND search is performed with “That” and “Tatsushi”.

  In this case, when “A” is input as the first word, the definite attribute is the facility name and the address name (see FIG. 5). Thereafter, when “ta” is input after the logical operator “&” is input, a registered word that is forward-matched with “ta” is read from the predicted word table. At this time, “Tatoshi” is also read out. However, “Tatusushi” has an attribute (facility name) that matches the fixed attribute (see FIG. 5). For this reason, the score value set to “TASTO” is subjected to subtraction processing using the score value set to “That”. Then, the correction score value of “addition” is “0 (zero)”. As a result, when there are N or more registered words that have been read out, the “assist” desired by the vehicle occupant is stored in the predicted word table because the correction score value is “0 (zero)”. Nevertheless, there is a possibility that guidance is not displayed on the display 13 as a prediction candidate.

  In this regard, in the present embodiment, as shown in FIG. 9, score values S1, S2, and S3 are set for each attribute in the registered word. Therefore, the score values S1, S2, and S3 of “Tatoshi” read because “ta” is input after the input of the logical operator “&” are the score values S1, S2, and S3 of “That”. Individually corrected. At this time, “That” and “Tatusushi” coincide with each other in that they have an attribute of facility name. As a result, the first score value S1 (100) of “addition” is corrected with the first score value S1 (348) of “yes”. On the other hand, “Tatusushi” has an attribute called “genre name”, whereas “Ayu” has no attribute called “genre name”. As a result, the third score value S3 of “addition” is not substantially corrected.

  Then, the total score value SHT of “addition” obtained by adding the correction score values SH1, SH2, and SH3 is not “0 (zero)”. Therefore, even when the number of read registered words is greater than N, the total score value SHT is not “0 (zero)”, so that “addition” desired by the vehicle occupant is predicted. The possibility of being displayed as guidance on the display 13 is increased.

Therefore, in this embodiment, in addition to the effects (1) to (4) of the first embodiment, the following effects can be obtained.
(6) In the predicted word table of the present embodiment, score values S1, S2, and S3 for each attribute are set for the registered word. Therefore, it is possible to increase the possibility that a registered word having another attribute that does not match the confirmed attribute among the registered words having an attribute that matches the confirmed attribute is guided and displayed on the display 13 as a prediction candidate. Therefore, it is highly possible that the registration word that may be selected by the vehicle occupant is displayed on the display 13 as a prediction candidate so as to contribute to shortening the time required for input.

(Third embodiment)
Next, a third embodiment of the present invention will be described with reference to FIGS. The third embodiment is different from the first and second embodiments in the contents of the prediction word table, the definite attribute determination processing routine, and a part of the prediction candidate determination processing routine. Therefore, in the following description, parts different from the first and second embodiments will be mainly described, and the same reference numerals are given to the same or corresponding member configurations as those of the first and second embodiments. A duplicate description will be omitted.

  As shown in FIG. 12, two types of predicted word tables are stored in the table storage unit 25 (see FIG. 1) of the present embodiment. In the present embodiment, unlike the second embodiment, some attributes are finely classified. For example, the address name is classified into a prefecture level address name, a city level address name, a ward level address name, and a town and village level address name. Among these attributes related to the area name, the address name at the town and village level can be said to be an attribute indicating an area in a narrow range as compared with other address names. It can be said that the address name at the ward level is an attribute indicating a wider area than the address name at the town and village level. It can be said that the address name at the city level is an attribute indicating a wider area than the address names at the town and village level and the ward level. The address name at the prefecture level can be said to be an attribute indicating a wider area than other address names.

  Genre names are classified into genre names at a large concept level, genre names at a type level, and genre names at a store name level. The genre name at the large concept level is an attribute for roughly classifying the genre. Examples of registered words having such a genre name at a high concept level include “Outlet (Outlet Mall)”, “Restaurant”, “Play (Play)”, “Eat (Eat)” (not shown), and the like.

  The genre name at the type level is an attribute that makes the genre name at the large concept level into a lower concept (invention). Examples of registered words subordinate to “restaurant” and “eat (eat)” include “ice (ice cream)”, “sushi (sushi)”, “hamburger” (not shown), and the like. Also, examples of registered words in which “play (play)” is subordinated include “Eiga (movie)”, “Yuenchi (amusement park)”, “karaoke” (not shown), and the like.

  The genre name at the store name level is an attribute that makes the genre name at the kind level into a lower concept (invention). That is, the genre name at the store name level is set to a registered word representing a specific store name, position, and the like. An example of a registered word that subordinates “sushi (sushi)” is “tastashi (food shop)”.

First, the first predicted word table shown on the upper side of FIG. 12 will be described.
In the first predicted word table, score values S1, S21, S22, S23, S24, S31, S32, and S33 are set for each attribute for each registered word. Specifically, regarding the registered word “ah” representing the company name, the first score value S1 for the facility name is “260”, while the other score values S21 to S24 and S31 to S33 are “0 (zero). ) ”. Regarding the ice cream abbreviation “ice”, the 32nd score value S32 for the genre name of the type level is “380”, and the other score values S1, S21 to S24, S31, S33 are “0 (zero)”. It is. Further, regarding the “name” representing the facility name or the address name at the prefecture level, the first score value S1 is “348” and the 21st score value S21 for the address name at the prefecture level is “232”. On the other hand, the other score values S22 to S24 and S31 to S33 are “0 (zero)”.

  Regarding the “Outlet” which is an abbreviation of the outlet mall, the 31st score value S31 for the genre name of the large concept level is “380”, while the other score values S1, S21 to S24, S32, and S33 are “0 (zero). ) ”. Regarding “Iai” representing the facility name or the address name at the prefecture level, the first score value S1 is “225” and the twenty-first score value S21 is “225”, while the other score values S22 to S24. , S31 to S33 are “0 (zero)”. Regarding “Oh,” which represents a facility name or a city level address name, the first score value S1 is “160” and the 22nd score value S22 for the city level address name is “240”, while others The score values S21, S23, S24, and S31 to S33 are “0 (zero)”.

  Regarding “sushi (sushi)” representing the type of food, the 32nd score value S32 for the genre name at the type level is “350”, while the other score values S1, S21 to S24, S31, S33 are “0 ( Zero) ”. With respect to “addition” representing the store name, the first score value S1 is “100” and the 33rd score value S33 for the genre name at the store name level is “100”, while the other score values S21 to S24. , S31 and S32 are “0 (zero)”. Regarding “Nanu” representing an address name at the town and village level, the 23rd score value S23 for the address name at the town and village level is “150”, while the other score values S1, S21 to S23, and S31 to S33 are “0”. (Zero) ".

  In this embodiment, “Ooi (City)” is the name of the city where the corresponding area has only one place in Japan, whereas “Nananu (City)” is the name of the area in Japan. It is assumed that the town name exists in multiple places.

Next, the second predicted word table shown on the lower side of FIG. 12 will be described.
Depending on the registered word corresponding to the confirmed character string, prediction candidates for the character string before confirmation input after the logical operator “&” may be limited. For example, when the word corresponding to the confirmed character string is “Ooi (city)”, there is only one place in the country. Therefore, it is unlikely that a word having an address name at the prefecture level is input after the logical operator “&”. When the word corresponding to the confirmed character string is “sushi (sushi)”, a word having a genre of a large conceptual level (for example, “restaurant”) is input after the logical operator (&). Unlikely. Depending on the words that have been confirmed in this way, prediction candidates for the next input word can be narrowed down. In order to narrow down such prediction candidates, a second prediction word table is prepared.

  As shown in FIG. 12, in the second predicted word table, exclusive score values Z1, Z21, Z22, Z23, Z24, Z31, Z32, and Z33 are set for each registered word for each attribute. Specifically, regarding the ice cream abbreviation “ice”, the 31st exclusive score value Z31 for the genre name of the large conceptual level is “200”, and the other exclusive score values Z1, Z21 to Z24, Z32, Z33 is “0 (zero)”. As for the facility name or the address name at the city level, the 21st exclusive score value Z21 for the prefecture level is “210”, while the other exclusive score values Z1, Z22 to Z24, Z31 to Z33 are “0 (zero)”.

  Regarding “sushi (sushi)” representing the type of food, the 31st exclusive score value Z31 for the genre name at the high concept level is “200”, while the other exclusive score values Z1, Z21 to Z24, Z32, Z33 are “0 (zero)”. With regard to “addition” representing the store name, the 31st exclusive score value Z31 and the 32nd exclusive score value Z32 for the genre names of the large concept level and the kind level are “100”, while the other exclusive score values Z1 and Z21. -Z24 and Z33 are "0 (zero)".

  On the other hand, the exclusive score values Z1, Z21 to Z24, Z31 to Z33 of other registered words are all “0 (zero)”. This is because when another registered word becomes a word corresponding to a confirmed character string, the prediction candidate of the word input following the logical operator “&” is selected from the attribute of the other registered word (determined attribute). ) Is not required to be excluded.

Next, the fixed attribute determination processing routine of this embodiment will be described with reference to the flowchart shown in FIG.
In the confirmed attribute determination processing routine, when a word corresponding to the confirmed character string is present in the predicted word table (step S110: YES), the control unit 22 calculates the score value S (S1, S21 to S21) of the searched registered word. S24, S31 to S33) are acquired from the first predicted word table (step S131). In the present embodiment, as in the case of the second embodiment, a score value is set for each attribute. Therefore, obtaining each score value of a registered word that matches the condition is equivalent to obtaining the attribute (attribute information) of the registered word. Therefore, in the present embodiment, step S131 corresponds to an attribute acquisition step.

  Subsequently, the control unit 22 acquires the exclusive score value Z (Z1, Z21 to Z24, Z31 to Z33) of the searched registered word from the second predicted word table (step S132). And the control part 22 sets the correction value H (H1, H21, H22, H23, H24, H31, H32, H33) for every attribute based on the score value S and the exclusive score value Z acquired by step S131, S132. (Step S133), and the definite attribute determination processing routine is terminated.

  On the other hand, when the word corresponding to the confirmed character string is not in the predicted word table (step S110: NO), the control unit 22 corrects the correction value H (H1, H21, H22, H23, H24, H31, H32) for each attribute. , H33) are all set to “0 (zero)” (step S134), and the definite attribute determination processing routine ends.

Here, the method of setting the correction value H will be described by taking as an example the case where the word corresponding to the confirmed character string is “Oh!”.
When the registered word retrieved is “Oh!”, The first score value S1 is “160”, the 22nd score value S22 is “240”, and the other score values S21, S23, S24, S31 to S33 are It becomes “0 (zero)” (see FIG. 12). The 21st exclusive score value Z21 is "210", and the other exclusive score values Z1, Z22 to Z24, and Z31 to Z33 are "0 (zero)".

  Since the first score value S1 is “160” and the first exclusive score value Z1 is “0 (zero)”, the first correction value H1 is set to “160”. Further, since the 21st score value S21 is “0 (zero)” and the 21st exclusive score value Z21 is “210”, the 21st correction value H21 is set to “210”. Further, since the 22nd score value S22 is “240” and the 22nd exclusive score value Z22 is “0 (zero)”, the 22nd correction value H22 is set to “240”. Further, since both the 23rd score value S23 and the 23rd exclusive score value Z23 are “0 (zero)”, the 23rd correction value H23 is set to “0 (zero)”. Similarly, the 24th correction value H24, the 31st correction value H31, the 32nd correction value H32, and the 33rd correction value H33 are each set to “0 (zero)”.

  If the first score value S1 is “160 (> 0 (zero))” and the first exclusive score value Z1 is “100 (> 0 (zero))”, the first correction value H1 may be the larger of the first score value S1 and the first exclusive score value Z1, or may be the average value of the first score value S1 and the first exclusive score value Z1.

Next, the prediction candidate determination processing routine of this embodiment will be described with reference to the flowchart shown in FIG.
In the prediction candidate determination process routine, the control unit 22 sequentially executes processes equivalent to the processes in steps S200, S201, S202, and S203. Subsequently, the control unit 22 arranges the registered words read out in step S201 in the order of the Japanese syllabary, and acquires each score value S (M) of the Mth registered word from the top from the predicted word table (step S241). And the control part 22 performs a correction process with respect to each score value S (M) of the acquired Mth registration word (step S242). That is, the control unit 22 calculates the score values S1 (M), S21 (M), S22 (M), S23 (M), S24 (M), S31 (M), S32 (M), and S33 (M), Correction is performed using the correction values H1, H21 to H24, H31 to H33 set in step S133 or step S134. And the control part 22 makes correction | amendment results SH1 (M), SH21 (M), SH22 (M), SH23 (M), SH24 (M), SH31 (M), SH32 (M), SH33 ( M). The correction process for the first score value S1 (M) will be described as an example. The control unit 22 subtracts the correction value H1 from the first score value S1 (M), and the subtraction result is used as the first correction score value SH1 (M ).

  Subsequently, the control unit 22 adds the correction score values SH1 (M), SH21 (M) to SH24 (M), and SH31 (M) to SH33 (M) calculated in step S242, and adds the total results. The score value SHT (M) is set (step S243). Then, similarly to the determination process in step S208, the control unit 22 determines whether or not the current coefficient M matches the total read number P acquired in step S202 (step S244). When the coefficient M is less than the total number P to be read (step S244: NO), the control unit 22 increments the coefficient M by “1” (step S245), and the process proceeds to step S241 described above. On the other hand, when the coefficient M matches the total number P to be read (step S244: YES), the control unit 22 sequentially executes the processes of steps S210 and S214, and ends the prediction candidate determination process routine.

  As mentioned above, “Ooi (City)” is a city name that has only one location in Japan, so if the word corresponding to the confirmed character string is “Ooi” It is unlikely that the word entered after the operator “&” is a prefecture name. In the present embodiment, regarding “Oh,” the 21st exclusive score value Z21 for the address name at the prefecture level is “210” (see FIG. 12).

  If the character entered after the logical operator “&” is “A”, all registered words that match the prefix “A” are “Ah”, “Ice”, “Ah”, “Outlet”, etc. Read out. Of the registered words read out in this way, “Ah”, “Ice”, and “Outlet” do not have a prefecture-level address name as an attribute. Therefore, the score value S set for these is not corrected by the 21st exclusive score value Z21 (210) of “Oh no”. On the other hand, “Ayu” has an address name at the prefecture level as an attribute. For this reason, the 21st score value S21 (225) for the address name at the prefecture level set to “Aoi” is corrected by the 21st exclusive score value Z21 (210) of “Aoi”. That is, the 21st corrected score value SH21 of “Any” is “15” (= 225-210).

  As a result, the combined score value SHT of “that” becomes low. As a result, it is less likely that “A” will be displayed on the display 13 as a prediction candidate for the character string “A” before confirmation. Even if the guidance is displayed, “A” is displayed in a lower order than the other prediction candidates.

  In addition, when the word corresponding to the confirmed character string is “addition”, the word input following the logical operator “&” is the attribute of the genre name at the type level or the genre name at the high concept level. Is unlikely to be a word with Therefore, in the present embodiment, regarding “addition”, the 31st exclusive score value Z31 for the genre name of the large concept level is set to “100”, and the 32nd exclusive score value Z32 for the genre name of the type level is set. Is set to “100” (see FIG. 12). As a result, even if, for example, “s” is input following the logical operator “&”, “sushi (sushi)” can be guided and displayed on the display 13 as a prediction candidate of the character string “s” before confirmation. The nature becomes low. Even if the guidance is displayed, “sushi” is displayed below the other prediction candidates.

Therefore, in this embodiment, in addition to the effects (1) to (4) and (6) of the first embodiment, the following effects can be obtained.
(7) When performing an AND search, a word having a more general attribute than the attribute (determined attribute) of the word corresponding to the confirmed character string may be input following the logical operator “&”. Is considered low. For example, it is unlikely that “sushi (sushi)” will be input after “tatashi” representing the sushi restaurant name is input. Therefore, in this embodiment, exclusive score values Z1, Z21 to Z24, and Z31 to Z33 are set for each registered word. Then, using this exclusive score value Z1, Z21 to Z24, Z31 to Z33, the score value S of the registered word that matches the character string before confirmation is corrected. As a result, it is possible to reduce the possibility that a registered word having a larger concept attribute than the word attribute (determined attribute) corresponding to the confirmed character string is guided and displayed on the display 13 as a prediction candidate. Therefore, it is possible to increase the possibility that a registered word that is highly likely to be selected by a vehicle occupant is guided and displayed on the display 13.

  (8) Also, registered words representing address names (region names) are classified into words having a plurality of corresponding regions and words having only one corresponding region. If the registered word corresponding to the region name having only one location in Japan is a word corresponding to the confirmed character string, the registered word representing a region wider than the region indicated by the registered word is The possibility of being displayed as guidance on the display 13 as a prediction candidate is low. For example, if the word corresponding to the confirmed character string is “Ooi (City)”, the 21st exclusive score value Z21 for the address name at the prefecture level is set to “210” for the “Ooi”. . Therefore, the possibility that the registered word representing the address name at the prefecture level is displayed on the display 13 as a prediction candidate is reduced. Therefore, it is possible to reduce the possibility that a registered word that is unlikely to be selected by a vehicle occupant is guided and displayed on the display 13.

(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described with reference to FIGS. Note that the fourth embodiment is different from the third embodiment in that a prediction candidate determination processing routine for NOT search is executed. Therefore, in the following description, parts different from those of the third embodiment will be mainly described, and the same or corresponding member configurations as those of the second embodiment are denoted by the same reference numerals and redundant description will be omitted. Shall.

First, an example of a search display screen displayed on the display 13 when performing a NOT search will be described with reference to FIG.
When a NOT search is performed with a plurality of words, the NOT region 45 is selected after the input of the first word (in this case, “sushi”) is completed. Then, as shown in FIG. 15, “−” is displayed in the selected character display area 41 as a NOT search logical operator following the word “sushi”. Then, the first word is fixed to “sushi”. That is, a character string (in this case, “sushi”) displayed before the logical operator “−” (left side in FIG. 15) corresponds to a confirmed character string.

  Then, when the input of the second word is started following the logical operator “-”, the character displayed on the right side of the logical operator “-” is displayed on the right side of the input-capable character area 40 on the display 13. A predicted word display area 46 in which prediction candidates estimated from the columns are displayed is formed. In this way, the character string displayed on the right side of the logical operator “−” corresponds to the character string before being confirmed. In the predicted word display area 46, at least one of the words that coincides with the character string before confirmation among the registered words stored in the predicted word table shown in FIG. 9 is displayed. For example, when the character string before confirmation is “ta”, the predicted word display area 46 displays a word that matches forward with “ta” as a prediction candidate.

  When any one of the plurality of prediction candidates displayed in the predicted word display area 46 (for example, “tassault”) is selected, the selected character display area 41 is a character string before confirmation. “Ta” is replaced with the selected prediction candidate (plus). As a result, “sushi-tassushi” is displayed in the selected character display area 41. When the search execution area 43 is selected in this state, a NOT search is performed on the first word (sushi) and the second word (tass). In this case, a restaurant that provides sushi other than “tatasushi” is searched.

  Next, the fixed attribute determination processing routine for NOT search will be described with reference to the flowchart shown in FIG. Note that the fixed attribute determination processing routine for NOT search is almost the same as the fixed attribute determination processing routine for AND search, except for the method of correcting the score value of the registered word that matches forward with the character string before determination. Therefore, the points different from the case of the AND search will be explained mainly, and the other parts will be explained briefly.

  This prediction candidate determination processing routine is executed every time a character is selected by the vehicle occupant after the logical operator “-” for NOT search is input. In addition, the prediction candidate determination processing routine is executed when the character string before determination exists even at the timing when the correction area 42 is selected by the occupant. Then, in the prediction candidate determination processing routine, when there is no registered word that matches the character string before confirmation in each registered word included in the predicted word table (step S400: NO), the control unit 22 selects the prediction candidate. The decision processing routine is terminated. On the other hand, when there is a registered word that matches forward with the character string before confirmation (step S400: YES), the control unit 22 adds the registered word that matches forward with the character string before confirmation and the attribute of the registered word from the predicted word table. All are read (step S401). Therefore, in this embodiment, step S401 corresponds to a reading step. Subsequently, the control unit 22 acquires the total number of registered words read from the predicted word table as the read total number P (step S402), and sets “1” to the coefficient M (step S403).

  Then, the control unit 22 arranges the registered words read out in step S401 in the order of the Japanese syllabary, and determines whether or not the Mth registered word from the top has a lower concept attribute of the definite attribute (step S404). For example, the attribute relating to the address name is classified into a prefecture level, a city level, a ward level, and a town and village level (see FIG. 12). In view of the attribute of the address name at the prefectural level, the attribute of the address name at the city level, the ward level, and the town level corresponds to the attribute of the subordinate concept. Similarly, from the viewpoint of the city level address name attribute, the city level and town level address name attributes correspond to subordinate concept attributes. Further, the attributes relating to the genre name are classified into a large concept level, a kind level, and a store name level. In view of the attribute of the genre name at the large concept level, the attribute of the genre name at the type level and the store name level corresponds to the attribute of the lower concept. Similarly, from the attribute of genre name at the type level, the attribute of genre name at the store name level corresponds to the attribute of the lower concept.

  In step S404, for example, when the word corresponding to the confirmed character string is “sushi (sushi)”, the confirmed attribute is the genre name of the type level. If the read registered word is “tassy”, the “tassy” has an attribute of a genre name at the store name level, which is an attribute of a subordinate concept of the definite attribute. Further, for example, when the word corresponding to the confirmed character string is “Yes”, the confirmed attribute is the facility name and the address name at the prefecture level. When the read registered word is “Nanu”, the “Nanu” has an attribute of an address name at the town / village level, which is an attribute of a subordinate concept of the deterministic attribute.

  If the Mth registered word does not have the attribute of the subordinate concept of the definite attribute (step S404: NO), the control unit 22 sets the score value S (M) of the Mth registered word as the Mth registered word. The correction score value SH (M) of the registered word is set (step S405), and the process proceeds to step S407 described later. On the other hand, when the Mth registered word has the attribute of the subordinate concept of the definite attribute (step S404: YES), the control unit 22 performs a correction process for increasing the score value S (M) of the Mth registered word ( Step S406). In the present embodiment, the control unit 22 adds a preset correction value H (for example, 50) to the score value S (M) of the Mth registered word, and uses the addition result as the correction score value SH (M). And Then, the control part 22 transfers the process to following step S407.

  If the current coefficient M is less than the total number P read out acquired in step S402 (step S407: NO), the control unit 22 increments the coefficient M by “1” (step S408), and the process is described above. The process proceeds to step S404. On the other hand, when the coefficient M coincides with the total number P to be read (step S407: YES), the control unit 22 determines that the determination process of step S404 has been performed for all registered words read from the predicted word table. As described above, when the determination result in step S407 is affirmative, among the registered words read from the predicted word table, the score value (in this case, the corrected score value) of the registered word having the attribute of the subordinate concept of the definite attribute. (SH) and the score value (in this case, the corrected score value SH) of other registered words that do not have the attribute of the subordinate concept of the definite attribute are changed.

  Then, the control unit 22 arranges the registered words read from the predicted word table in descending order of the correction score value SH, and extracts the top N registered words (step S409). Subsequently, the control unit 22 performs a prediction candidate display process for guiding and displaying the N registered words extracted in step S409 on the display 13 as prediction candidates (step S412), and then ends the prediction candidate determination processing routine. . In this case, among each registered word read out from the predicted word table, a registered word having a lower concept attribute than a confirmed attribute has priority over other registered words having no lower concept attribute than a confirmed attribute. The guidance is displayed on the display 13. Therefore, in this embodiment, a prediction display step is comprised by step S404-S412.

Therefore, in this embodiment, in addition to the effects (1) to (4) and (6) to (8) of the first embodiment, the following effects can be obtained.
(9) When a NOT search is performed, the attribute of the word input following the NOT search logical operator “-” has a lower concept than the attribute of the word (determined attribute) corresponding to the confirmed character string. It is likely to be an attribute. Therefore, in this embodiment, when a NOT search is performed, the priority (in this case, the score value SH) of a registered word having a lower concept attribute than the fixed attribute is increased. On the other hand, the priority (in this case, the score value SH) of a registered word that does not have a lower-level concept attribute than the definite attribute is maintained. After performing such correction processing, among the read registered words, the registered word having a higher priority is guided to the display 13 as a character string prediction candidate before being determined in preference to the registered word having a lower priority. Is displayed. Therefore, when performing a NOT search, it is possible to improve the prediction accuracy of a character string being input. Therefore, when a NOT search is performed, there is a high possibility that the time required for input can be reduced.

In addition, you may change said each embodiment into another embodiment as follows.
-In 2nd Embodiment, you may change a prediction word table into a table as shown in FIG. That is, as shown in FIG. 17, each registered word is associated with a score value S and an attribute ratio (%) for each attribute. Then, for example, when “Ooh” is read from the predicted word table, the first score value S1 for the facility name of “Ooh” is the value of “40%” of the score value S (400) (ie, 160). ) And the second score value S2 for the address name “Oh!” Is set to a value of “60%” of the score value S (400) (ie, 240). Therefore, even when the predicted word table is configured as shown in FIG. 17, the same effects as those of the second embodiment can be obtained. Note that the prediction word table (FIG. 12) of the third embodiment may have the same configuration as the table shown in FIG.

  In the third and fourth embodiments, the genre names at the large concept level may be subdivided. For example, the genre name at the large conceptual level is the genre name related to behavior such as “eat (eat)” and “play (play)”, and the genre name related to facilities such as “restaurant” and “Kissaten (coffee shop)”. May be classified. In this case, the genre name related to the facility corresponds to a subordinate attribute of the genre name related to the action.

  In each embodiment, when the read total number P acquired in step S202 is less than N, all the read registered words may be guided and displayed on the display 13 as prediction candidates.

  In each embodiment, in step S201, a registered word that partially matches the character string before confirmation may be read from the predicted word table. Here, “partial match” is a concept including forward match and backward match.

  In the first embodiment, it may be prohibited to guide and display a registered word having an attribute that matches the confirmed attribute as a prediction candidate on the display 13. For example, a flag is individually set for each registered word read from the predicted word table, and a flag of a registered word having an attribute that matches the confirmed attribute is turned off, while a registration that does not have an attribute that matches the confirmed attribute The word flag is turned on. Then, a registered word whose flag is on may be guided and displayed on the display 13 as a prediction candidate. Even in this case, when the number of registered words whose flag is turned on is less than N, the registered word whose flag is turned off may be guided and displayed on the display 13 as a prediction candidate.

  -In each embodiment, at the time of AND search, you may perform the correction process which raises the score value S of the registration word which does not have the attribute which corresponds with a definite attribute. For example, in the first embodiment, when the determination result in step S205 is negative, the score value S of the Mth registered word is added with the correction value H, and the addition result is set as the correction score value SH (M). . When the determination result of step S205 is affirmative, the score value S of the Mth registered word is set as a corrected score value SH (M).

  In each embodiment, the maximum number of registered words that are guided and displayed on the display 13 as prediction candidates is N. However, all registered words that are forward-matched with the character string before confirmation are guided and displayed on the display 13. Also good. However, it is preferable to display a registered word having a high priority (corrected score value SH) after the correction process higher than a registered word having a low priority after the correction process.

  In each embodiment, N, which is the number of registered words that are guided and displayed on the display 13 as prediction candidates, may be an arbitrary number other than “5” (for example, 3) as long as it is an integer of 1 or more.

  In each embodiment, the score value S of the predicted word table may be learned as appropriate. For example, when a registered word displayed as a prediction candidate on the display 13 is selected by a vehicle occupant, the score value S of the registered word may be increased by a predetermined value (for example, 10) set in advance. . If comprised in this way, the score value S of the registration word with many selected numbers will become high compared with the score value S of another registration word. Therefore, it becomes possible to display guidance candidates on the display 13 according to the taste of the vehicle occupant.

  In each embodiment, the ROM 22b of the control unit 22 may also serve as the table storage unit 25 that stores the prediction word table if the contents of the prediction word table are not changed. That is, the predicted word table may be stored in the ROM 22b. In this case, the ROM 22b functions as a storage unit.

  In each embodiment, prediction candidates may be guided and displayed on the display 13 using the prediction word table even when the first word is input. In this case, a registered word that matches the character string in the middle of input is read from the predicted word table, and at least one of the read registered words is guided and displayed in the predicted word display area 46 of the display 13. And the prediction candidate selected by the passenger | crew among each prediction candidate guided and displayed on the prediction word display area 46 may be displayed on the selected character display area 41 as the first word. With this configuration, it is possible to expect further reduction in the time required for input.

  The point search method of the present invention may be embodied when performing an AND search using an arbitrary number of three or more words (for example, three). For example, when performing an AND search with three words, when the second word is input, the character string indicating the first word is set as the fixed character string, and the attribute of the word is set as the fixed attribute. The Then, a registered word to be guided and displayed on the display 13 as a prediction candidate is selected based on the confirmed attribute. Subsequently, when the third word is input, the character string indicating the first word and the character string indicating the second word are determined character strings, and the attributes of the respective words are determined attributes. The Then, a registered word to be guided and displayed on the display 13 as a prediction candidate is selected based on the confirmed attribute.

  -In each embodiment, selection of the prediction candidate guided and displayed on the display 13 may be a selection operation of a remote controller by a vehicle occupant, or may be selection using voice input by a vehicle occupant.

  The point search device of the present invention may be embodied in a mobile information terminal such as a mobile phone. In this case, a point search program for causing the mobile information terminal to execute the processing routines is installed, and the processing routines are appropriately executed.

  In addition, the point search device of the present invention may be embodied in an information terminal such as a personal computer. In this case, the point search program is installed in the information terminal, and the above processing routines are appropriately executed.

Next, technical ideas that can be grasped from the above embodiments and other embodiments will be added below.
(A) A point search device for searching for point candidates that match the conditions indicated by a plurality of input character strings,
Storage means for storing a plurality of registered character strings and attribute information indicating attributes of the respective registered character strings;
Attribute registration means for acquiring attribute information of the registered character string as attribute information of the confirmed character string when a registered character string corresponding to the confirmed character string is stored in the storage means;
A reading means for reading out from the storage means a registered character string that at least partially matches a character string before confirmation that is input following the confirmed character string, and attribute information of the registered character string;
When performing a NOT search with a plurality of character strings, among the registered character strings read by the reading means, a registered character string having attribute information of a lower concept than the attribute information acquired by the attribute acquiring means, Prediction display means for displaying on the display screen as a prediction candidate of the character string before the determination in preference to the registered character string not having the attribute information of the lower concept,
When the prediction candidate displayed on the display screen is selected by the prediction display unit, a replacement display unit that replaces the character string before confirmation with the prediction candidate and displays it on the display screen is provided. A point search device.

(B) A point search method for searching for point candidates that match the conditions indicated by a plurality of input character strings,
When a registered character string corresponding to a confirmed character string is stored in a storage unit that stores a plurality of registered character strings and attribute information indicating the attribute of each registered character string, the attribute information of the registered character string is An attribute acquisition step of acquiring from the storage means as attribute information of a confirmed character string;
A read step for reading from the storage means a registered character string that at least partially matches a character string before confirmation that is input subsequent to the confirmed character string, and attribute information of the registered character string;
When performing a NOT search with a plurality of character strings, among the registered character strings read out in the reading step, a registered character string having lower-level attribute information than the attribute information acquired in the attribute acquisition step is used. A prediction display step of displaying on the display screen as a prediction candidate of the character string before the determination with priority over the registered character string not having the attribute information;
A replacement display step of, when a user selects a prediction candidate displayed on the display screen in the prediction display step, replacing the pre-determined character string with the prediction candidate and displaying it on the display screen. The point search method.

(C) A point search program for causing the control device to search for point candidates that match the conditions indicated by the plurality of input character strings,
In the control device,
Storage means for storing a plurality of registered character strings and attribute information indicating attributes of the respective registered character strings;
Attribute registration means for acquiring attribute information of the registered character string as attribute information of the confirmed character string when a registered character string corresponding to the confirmed character string is stored in the storage means;
A reading means for reading out from the storage means a registered character string that at least partially matches a character string before confirmation that is input following the confirmed character string, and attribute information of the registered character string;
When performing a NOT search with a plurality of character strings, among the registered character strings read by the reading means, a registered character string having attribute information of a subordinate concept of the attribute information acquired by the attribute acquiring means Prediction display means for displaying on the display screen as a prediction candidate of the character string before the determination with priority over a registered character string that does not have conceptual attribute information;
When a prediction candidate displayed on the display screen is selected by the prediction display unit, the prediction display unit functions as a replacement display unit that replaces the character string before confirmation with the prediction candidate and displays it on the display screen. Point search program.

  DESCRIPTION OF SYMBOLS 12 ... Display apparatus, 13 ... Display as an example of a display screen, 20 ... Control apparatus as an example of a point search apparatus, 22 ... Attribute acquisition means, Read-out means, prediction display means, Control part as an example of replacement display means, 22b: ROM as an example of storage means; 25: Table storage unit as an example of storage means.

Claims (9)

A point search device that searches for point candidates that match the conditions indicated by a plurality of input character strings,
Storage means for storing a plurality of registered character strings and attribute information indicating attributes of the respective registered character strings;
Attribute registration means for acquiring attribute information of the registered character string as attribute information of the confirmed character string when a registered character string corresponding to the confirmed character string is stored in the storage means;
A reading means for reading out from the storage means a registered character string that at least partially matches a character string before confirmation that is input following the confirmed character string, and attribute information of the registered character string;
When performing AND search on a plurality of character strings, among registered character strings read by the reading means, match registered character strings that do not have attribute information that matches attribute information acquired by the attribute acquiring means. Prediction display means for displaying on the display screen as a prediction candidate of the character string before the determination with priority over a registered character string having attribute information to be
When a prediction candidate displayed on the display screen is selected by the prediction display means, a replacement display means for replacing the character string before the confirmation with the prediction candidate and displaying it on the display screen ,
The storage means stores the plurality of registered character strings, the attribute information for each registered character string, and the prediction priority set for each registered character string in association with each other,
The prediction display means includes
When performing AND search with multiple character strings,
Performing a correction process for correcting the priority of the registered character string based on the attribute information acquired by the attribute acquisition means;
Of the registered character strings read out from the storage unit by the reading unit, the registered character string having a high priority after the correction process is determined in preference to the registered character string having a low priority after the correction process. A point search device that displays on the display screen as a prediction candidate of a previous character string . Before Symbol prediction display means,
When performing AND search with multiple character strings,
Location search device according to claim 1, characterized in that intends row correction processing to lower the priority of the registered character string having the attribute information matching the attribute information acquired by the attribute acquisition unit.   The prediction display means includes
  When performing AND search with multiple character strings,
  The point search apparatus according to claim 1, wherein correction processing is performed to increase a priority of a registered character string having attribute information that does not match the attribute information acquired by the attribute acquisition unit. The prediction display means includes
When performing AND search with a plurality of character strings, a registered character string that does not have higher-level attribute information than the attribute information of the confirmed character string acquired by the attribute acquisition unit is used as the attribute information of the higher-level concept. 4. The point search device according to claim 1 , wherein the point search device is displayed on the display screen as a prediction candidate of the character string before the determination in preference to a registered character string having a character string. . The prediction display means includes
When performing an AND search with multiple character strings,
When the attribute information of the confirmed character string is attribute information related to the region name and there is only one candidate for the region indicated by the confirmed character string, the attribute related to the region name in a wider area than the region A registered character string that does not have information is displayed on the display screen as a prediction candidate of the character string before the determination, preferentially over a registered character string that has attribute information related to regional names of the wide area. The point search device according to any one of claims 1 to 4 . The prediction display means includes
When performing a NOT search with a plurality of character strings, among the registered character strings read by the reading means, a registered character string having attribute information of a lower concept than the attribute information acquired by the attribute acquiring means, any one of claims 1 to 5, characterized in that than registered character strings with no attribute information of a lower concept is displayed on the display screen as the predictive candidates preferentially the determined previous string The point search device described in the item. The storage means stores the plurality of registered character strings, the attribute information for each registered character string, and the prediction priority set for each registered character string in association with each other,
The prediction display means includes
When performing a NOT search with multiple character strings,
Performing a correction process for lowering the priority of the registered character string having the attribute information of the lower concept than the attribute information of the confirmed character string acquired by the attribute acquisition means,
Of the registered character strings read out from the storage unit by the reading unit, the registered character string having a high priority after the correction process is determined in preference to the registered character string having a low priority after the correction process. The point search device according to claim 6 , wherein the point search device is displayed on the display screen as a prediction candidate of the previous character string. A point search method by a point search device provided with a control unit that searches for point candidates that match the conditions indicated by a plurality of input character strings,
The controller is
When a registered character string corresponding to a confirmed character string is stored in a storage unit that stores a plurality of registered character strings and attribute information indicating the attribute of each registered character string, the attribute information of the registered character string is An attribute acquisition step of acquiring from the storage means as attribute information of a confirmed character string;
A read step for reading from the storage means a registered character string that at least partially matches a character string before confirmation that is input subsequent to the confirmed character string, and attribute information of the registered character string;
Attributes that match registered character strings that do not have attribute information that matches the attribute information acquired in the attribute acquisition step among the registered character strings read in the read step when performing an AND search with a plurality of character strings A prediction display step for displaying on the display screen as a prediction candidate of the character string before the determination with priority over a registered character string having information;
Wherein if the predicted display user predictive candidates displayed on the display screen in step is selected, have a, a replacement display step of displaying on the display screen a character string before confirm replaced by the predicted candidate,
The storage means stores the plurality of registered character strings, the attribute information for each registered character string, and the prediction priority set for each registered character string in association with each other,
In the prediction display step,
When performing AND search with multiple character strings,
Performing a correction process for correcting the priority of the registered character string based on the attribute information acquired in the attribute acquisition step,
Among the registered character strings read from the storage means in the reading step, the registered character string having a high priority after the correction process is preferentially determined over the registered character string having a low priority after the correction process. A point search method , comprising: displaying on the display screen as a prediction candidate of a previous character string . A point search program for causing a control device to search for point candidates that match the conditions indicated by a plurality of input character strings,
Said control device,
Storage means for storing a plurality of registered character strings , attribute information indicating an attribute of each registered character string, and a prediction priority set for each registered character string in association with each other ;
Attribute registration means for acquiring attribute information of the registered character string as attribute information of the confirmed character string when a registered character string corresponding to the confirmed character string is stored in the storage means;
A reading means for reading out from the storage means a registered character string that at least partially matches a character string before confirmation that is input following the confirmed character string, and attribute information of the registered character string;
When performing AND search on a plurality of character strings, among registered character strings read by the reading means, match registered character strings that do not have attribute information that matches attribute information acquired by the attribute acquiring means. Prediction display means for displaying on the display screen as a prediction candidate of the character string before the determination with priority over a registered character string having attribute information to be
When the prediction candidate displayed on the display screen by the prediction display means is selected, function as replacement display means for replacing the character string before confirmation with the prediction candidate and displaying on the display screen ,
When the control device performs an AND search with a plurality of character strings as the prediction display means,
While performing a correction process for correcting the priority of the registered character string based on the attribute information acquired by the attribute acquisition means,
Of the registered character strings read out from the storage unit by the reading unit, the registered character string having a high priority after the correction process is determined in preference to the registered character string having a low priority after the correction process. A point search program that is displayed on the display screen as a prediction candidate of a previous character string .

Images