WO2016004893A1 - 一种文本信息的输入方法和装置 - Google Patents

一种文本信息的输入方法和装置 Download PDF

Info

Publication number
WO2016004893A1
WO2016004893A1 PCT/CN2015/083756 CN2015083756W WO2016004893A1 WO 2016004893 A1 WO2016004893 A1 WO 2016004893A1 CN 2015083756 W CN2015083756 W CN 2015083756W WO 2016004893 A1 WO2016004893 A1 WO 2016004893A1
Authority
WO
WIPO (PCT)
Prior art keywords
text information
screen
appearance frequency
candidate
text
Prior art date
Application number
PCT/CN2015/083756
Other languages
English (en)
French (fr)
Inventor
任尚昆
张扬
Original Assignee
北京搜狗科技发展有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京搜狗科技发展有限公司 filed Critical 北京搜狗科技发展有限公司
Priority to JP2016564625A priority Critical patent/JP6667452B2/ja
Priority to KR1020167031676A priority patent/KR102327418B1/ko
Publication of WO2016004893A1 publication Critical patent/WO2016004893A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • G06F3/0233Character input methods
    • G06F3/0237Character input methods using prediction or retrieval techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • G06F3/0233Character input methods
    • G06F3/0236Character input methods using selection techniques to select from displayed items

Definitions

  • the embodiment of the present invention claims the priority of the Chinese patent application filed on July 11, 2014, the Chinese Patent Office, the application number is 201410332496.2, and the invention name is "a method and device for inputting text information", the entire contents of which are incorporated by reference. In the embodiment of the invention.
  • the present invention relates to the technical field of input methods, and in particular to a method for inputting text information and an input device for text information.
  • the technical problem to be solved by the embodiments of the present invention is to provide a method for inputting text information to improve input efficiency.
  • the embodiment of the invention further provides an input device for text information, which is used to ensure the implementation and application of the above method.
  • an embodiment of the present invention discloses a method for inputting text information, including:
  • the first text information includes all the key input strings corresponding to the on-screen text information, all of which are on Screen text information and the number of backspaces used for deletion;
  • the second text information includes text information that is re-screened after the first text information appears in the input record history;
  • the second text information is displayed in a specified location.
  • the step of calculating, according to the first text information, the selection probability that the second text information is selected on the screen comprises:
  • the selection probability that the candidate second text information is selected on the screen is calculated by using the following formula:
  • Py-Ac is the first text information
  • Py is a key input string corresponding to all the upper screen text information
  • A is the all-on-screen text information
  • c is The number of backspaces
  • B i is the second text information
  • count(Py-Ac) is the first historical appearance frequency
  • count(Py-AcB i ) is the second historical appearance frequency
  • n is the The number of candidate second text information
  • i is a positive integer.
  • the first historical appearance frequency and the second historical appearance frequency are obtained by:
  • All screened text will be displayed when it is detected that all or part of the on-screen text information has been deleted.
  • the key input string corresponding to the information, all the on-screen text information, and the number of backspaces used for deletion are recorded as the first text information;
  • the second historical appearance frequency of the second text information occurs after the occurrence of the first text information.
  • the step of determining the candidate second text information to be displayed in the second text information comprises:
  • the appearance frequency includes a ratio of the second historical appearance frequency to the first historical appearance frequency
  • the step of displaying the second text information in the specified location comprises:
  • the candidate second text information is displayed in the specified location.
  • the embodiment of the invention further discloses an input device for text information, comprising:
  • a text information determining module configured to determine the first text information and the second text information when detecting that all or part of the already-on-screen text information is deleted; wherein the first text information includes all the corresponding on-screen text information Key input string, all on-screen text information, and number of backspaces used for deletion; the second text information includes text information that is re-screened after the first text information appears in the input record history;
  • the selection probability calculation module is configured to calculate, according to the first text information, a selection probability that the second text information is selected on the screen;
  • a display module configured to display the second text information in the specified location when the selection probability is greater than a preset probability threshold.
  • the selection probability calculation module comprises:
  • a first searching module configured to search for a first historical appearance frequency of the first text information
  • a second search module configured to search for the occurrence of the first text message The second historical frequency of occurrence of the second text message
  • a candidate determining module configured to determine, in the second text information, candidate second text information to be displayed
  • the formula calculation module is configured to calculate a selection probability that the candidate second text information is selected on the screen by using the following formula:
  • Py-Ac is the first text information
  • Py is a key input string corresponding to all the upper screen text information
  • A is the all-on-screen text information
  • c is The number of backspaces
  • B i is the second text information
  • count(Py-Ac) is the first historical appearance frequency
  • count(Py-AcB i ) is the second historical appearance frequency
  • n is the The number of candidate second text information
  • i is a positive integer.
  • the device further comprises:
  • the first recording module is configured to: when it is detected that all or part of the on-screen text information is deleted, the key input string corresponding to all the on-screen text information, all the on-screen text information, and the number of backspace records used for deletion For the first text message;
  • a second recording module configured to record the information of the text that is re-screened after the occurrence of the first text information is the second text information
  • a first statistic module configured to count a first historical appearance frequency of the first text information
  • the second statistic module is configured to count the second historical appearance frequency of the second text information after the occurrence of the first text information.
  • the candidate determining module comprises:
  • a frequency calculation sub-module configured to calculate an appearance frequency of the second text information; the appearance frequency includes a ratio of the second historical appearance frequency to the first historical appearance frequency;
  • an extraction submodule configured to extract the one or more second text information having the highest frequency of occurrence as the candidate second text information displayed in the first screen.
  • the display module comprises:
  • a candidate display sub-module configured to display the candidate second text message in a specified location interest.
  • the embodiment of the invention further discloses a computer program comprising computer readable code, when the computer readable code is run on the terminal, causing the terminal to execute the aforementioned input method of the text information.
  • the embodiment of the invention also discloses a computer readable medium, wherein the aforementioned computer program is stored.
  • the embodiments of the invention include the following advantages:
  • the first text information and the second text information are determined, and the second text information is calculated according to the first text information, and the screen is selected on the first screen. Selecting the probability, when the selection probability is greater than the preset probability threshold, displaying the second text information in the first screen, fully utilizing the attribute of the text information re-code, and the feedback information of the user deleting the text information after the screen is uploaded, by using the user After the backspace, the text information of the upper screen will be associated with the text, and the text information that may be input is directly displayed, thereby avoiding the user repeatedly inputting the key input string, turning the page, etc., on the one hand, effectively reducing the user's key rate and improving Input efficiency, fast and convenient input, improve user experience, on the other hand, can reduce the response of the device to the user input key input string, page turning, etc., reducing the consumption of equipment system resources.
  • the text information that may be input after the backspace behavior is sorted according to the probability from high to low, which is convenient for the user to select the text information again, thereby improving the simplicity of the operation.
  • FIG. 1 is a flow chart showing the steps of an embodiment of a method for inputting text information according to the present invention
  • FIG. 2 is a diagram showing an example of a flow of normally inputting text information
  • FIG. 3 is a diagram showing an example of a process of inputting text information in a technique of restoring a scene first;
  • FIG. 4 is a diagram showing an example of an input flow of text information according to the present invention.
  • FIG. 5 is a diagram showing an example of a flow of normally inputting text information
  • FIG. 6 is a diagram showing an example of an input flow of text information according to the present invention.
  • FIG. 7 is a diagram showing an example of a flow of normal input text information
  • FIG. 8 is a diagram showing an example of a flow of inputting text information in a technique of restoring a scene first;
  • FIG. 9 is a diagram showing an example of an input flow of text information according to the present invention.
  • FIG. 10 is a structural block diagram of an embodiment of an input device for text information according to the present invention.
  • Figure 11 is a block diagram showing the structure of a terminal for executing an input method of text information according to an embodiment of the present invention
  • Figure 12 is a storage unit of the present invention for holding or carrying program code for implementing an input method of text information according to an embodiment of the present invention.
  • the user often selects the text information to be screened due to an input error or an error, and has to re-delete the mis-transmission portion and re-enter the input, thereby reducing input efficiency and reducing the user input experience.
  • the embodiment of the present invention uses the user's backspace behavior to analyze and the user needs
  • the text information of the re-screening is provided to the user, thereby effectively utilizing the input feedback information of the user to improve the user input efficiency, that is, the embodiment of the present invention can provide an active association of the text information when the backspace deletion behavior occurs after the screen is displayed, and is based on The analysis of the user's backspace behavior, when the user deletes some text information, actively displays the text information that the user is likely to continue to input in the first screen.
  • one of the core concepts of the foregoing embodiments of the present invention may be applied to the case where the text information is on the screen, and may also be applied to the case where the text information is not on the screen, which is not limited by the embodiment of the present invention.
  • FIG. 1 a flow chart of steps of an embodiment of a method for inputting text information according to the present invention is shown, which may include the following steps:
  • Step 101 When it is detected that all or part of the on-screen text information is deleted, determining the first text information and the second text information;
  • the first text information may include a key input string corresponding to all the on-screen text information, all the on-screen text information, and a number of backspaces used for deletion; the second text information may include the input record history. a text message that is re-screened after the first text message appears interest;
  • the upper screen can select the behavior of the candidate text information after the user inputs the key input string to obtain the candidate text information, and the backspace can delete the input text information for the user.
  • the key input string may be a key input string corresponding to an input mode such as pinyin, five strokes, a stroke, and an English.
  • the embodiment of the present invention does not limit this; the text information may be a text, a number, or a The symbol "", "%” and the like are not limited in this embodiment of the present invention.
  • the user may delete all or part of the currently on-screen text information by using a backspace event, and the backspace event may be triggered by a backspace key.
  • the backspace key also known as the backspace key, can be used to move the cursor one space to the left and delete the character at the left of the cursor.
  • the backspace key on the keyboard is labeled " ⁇ " and the English letter is BackSpace.
  • the first text information Py-Ac appearing in the input record history may be input in advance (Py is a key input string corresponding to all the upper screen text information, and A is the all-screen text information. c is the number of backspaces and its corresponding second text information Bi is stored in a database.
  • the first text information Py-Ac may be triple data, and the second text information may be an event that occurs under the condition that the triple data appears, and may form quad data Py-AcB with the triple data. i .
  • the text information may have a re-code in the input process, that is, the key input string form of the two candidate text information is the same, for example, the 26 keys in the keyboard are all in the same form (such as "I” and " The key string of the pinyin spelling is "wo"), and the 26 keys in the real keyboard are in the same form (such as the "master” and “director” pinyin spelling string is "zr "), the nine buttons in the Jiugongge keyboard are all in the same form (such as the “results” and “married” full-press keystroke strings are "543486"), the nine buttons in the Jiugongge keyboard are in the same form (such as "master” and “director” Pinyin spelling key string is "97") and so on.
  • the key input string Py corresponding to all the on-screen text information can be used to determine the re-code.
  • the user inputs a key input string of "woyou" in the environment of 26 keys in the keyboard. I have “I have” on the screen.
  • the text information of the recode has "I am” and so on, but there is no "I go”; for example, the key input is input in the environment of 9 buttons in the 9-square grid keyboard.
  • the string is "96968", and the text information of the re-code may be "I am going", so the text information of Lenovo may be different after the backspace.
  • the key input string Py corresponding to all the currently-on-screen text information may be determined by deleting all or part of the currently-on-screen text information, and all the currently-on-screen text information is A, currently deleting the number of backspace c used, obtaining the triplet data Py-Ac, that is, the first text information, and then searching for all the quad data of the Py-AcB i according to the triplet data, thereby determining the first Two text information B i .
  • all of the on-screen text information A and the second text information B i may be the same. For example, many users will continue to input "I” after deleting "I".
  • Step 102 Calculate, according to the first text information, a selection probability that the second text information is selected on the screen;
  • the first screen may be an input method interface for selecting candidate text information without performing candidate operations such as page turning and sliding.
  • the first text information there may be multiple second text information acquired from the history input record saved in the database, and each of the second text information is respectively calculated according to the first text information in the input record history. Select the selection probability of the upper screen.
  • step 102 may include the following sub-steps:
  • Sub-step S11 searching for a first historical appearance frequency of the first text information
  • Sub-step S12 searching for the second historical appearance frequency of the second text information after the occurrence of the first text information
  • the first historical appearance frequency may be an event in which the first text information appears.
  • the number of occurrences in the recording history, that is, the first historical frequency is the input of the key input string corresponding to all the on-screen text information, and the selection of all the on-screen text information is performed on the screen, and the execution is performed.
  • the frequency of the occurrence of the grid backspace in the history of the record; the second history frequency may be the number of times the event of the second text message appears after the first text message appears in the record history, that is,
  • the second historical frequency is a key input string corresponding to all the input upper screen text information, selecting all the upper screen text information upper screens, and performing the backspace of the backspace times, and then selecting the The behavior of the screen on which the text information is re-screened appears in the history of the record.
  • the first historical appearance frequency and the second historical appearance frequency may be obtained by:
  • Sub-step S21 when it is detected that all or part of the on-screen text information is deleted, the key input string corresponding to all the on-screen text information, all the on-screen text information, and the number of backspaces used for deletion are recorded as the first Text information
  • Sub-step S22 recording second text information that is re-screened after the first text information appears
  • Sub-step S23 counting the first historical appearance frequency of the first text information
  • Sub-step S24 counting the second historical appearance frequency of the second text information after the occurrence of the first text information.
  • first text information and the corresponding first historical appearance frequency, the second text information, and the corresponding second historical appearance frequency are saved as an input recording history in a database.
  • the habit of inputting text information of each user is different, so in order to meet the user's personalized needs, the probability of the second text information hitting the user's demand is further improved, and the first historical appearance frequency and the second historical appearance frequency may be The first historical appearance frequency of the first text information input for the recorded current user and the second historical appearance frequency of the second text information of the current user after re-screening the first text information.
  • the first text information and the second text information input by the same user in different situations may have unevenness, for example, the text information of the chat class is large, but the text message of the shopping class The information may be less, and the problem of sparse data is prone to occur. Therefore, in order to avoid data sparseness, the first historical frequency of occurrence may be determined based on the first text information input by all users in the present application, or may be based on all users. The second history occurrence frequency is determined by inputting the first text information and then re-screening the second text information.
  • the first historical appearance frequency may be the historical appearance frequency of the first text information input by the recorded entire users.
  • the second historical appearance frequency may be the historical appearance frequency of the second text information that is re-screened by the entire user of the record after inputting the first text information.
  • the first historical appearance frequency may also be a historical appearance frequency of the first text information input by the recorded all users, and the historical appearance frequency of the first text information input by the current user is mixed according to a certain ratio.
  • the second historical appearance frequency may also be the historical appearance frequency of the second text information after all the users input the first text information, and the history of the second text information is re-screened after the current user inputs the first text information. Frequency, mixed according to a certain proportion.
  • the embodiment of the invention does not limit this.
  • the first text information and the second text information input by the current user are not separately considered, but the information input by the entire user is more uniform, and the information input by the current user can be avoided. Unevenness results in a situation in which the first historical appearance frequency and the second historical appearance frequency are inaccurate.
  • the training process of the backspace association is entered. That is, counting the first historical appearance frequency count(Py-Ac) of the first text information and the second historical appearance frequency count(Py-AcB i ) of the second text information after the occurrence of the first text information, otherwise the user is considered
  • the text information is normally input, and the training process of the backspace association is not performed.
  • Sub-step S13 determining, in the second text information, candidate second text information to be displayed
  • the sub-step S13 may further include the following Substeps:
  • Sub-step S131 calculating an appearance frequency of the second text information; the appearance frequency may include a ratio of the second historical appearance frequency to the first historical appearance frequency;
  • Sub-step S132 extracting one or more second text information with the highest frequency of occurrence as the candidate second text information displayed in the first screen.
  • the frequency of occurrence of the quaternion data may be sorted from high to low, and then the number of second text information that the first screen can accommodate, for example, n, n is a positive integer, and then in the sorting.
  • the first n second text information is extracted as the candidate second text information displayed in the first screen.
  • the text information that may be on the screen after the backspace behavior is sorted according to the probability from high to low, which is convenient for the user to select the text information again, thereby improving the simplicity of the operation.
  • Sub-step S14 calculating the selection probability that the candidate second text information is selected on the screen by using the following formula:
  • Py-Ac is the first text information
  • Py is a key input string corresponding to all the upper screen text information
  • A is the all-on-screen text information
  • c is The number of backspaces
  • B i is the second text information
  • count(Py-Ac) is the first historical appearance frequency
  • count(Py-AcB i ) is the second historical appearance frequency
  • n is the The number of candidate second text information
  • i is a positive integer.
  • Step 103 When the selection probability is greater than a preset probability threshold, displaying the second text information at a specified location.
  • the specified location may be the first screen, or may be an association area (for example, an interface that generates a shape such as a bubble in the vicinity of the first screen), or may be an arbitrary area of the current interface of the device, which is not limited by the embodiment of the present invention. .
  • step 103 may comprise the following sub-steps:
  • Sub-step S31 displaying the candidate second text information at the specified position.
  • the selection probability p is greater than the probability threshold, indicating that the probability exceeds a certain probability and can be displayed on the first screen.
  • the text information of the user who wants to go on the screen next may be displayed, and the candidate second text information may be displayed on the first screen; if the selection probability p is less than or equal to the probability threshold, it indicates that the second text information can be hit in the first screen. If the probability that the text information of the user is going to be on the screen is too low, the second text information of the candidate can be discarded on the first screen.
  • the first text information and the second text information are determined, and the second text information is calculated according to the first text information, and the screen is selected on the first screen. Selecting the probability, when the selection probability is greater than the preset probability threshold, displaying the second text information in the first screen, fully utilizing the attribute of the text information re-code, and the feedback information of the user deleting the text information after the screen is uploaded, by using the user After the backspace, the text information of the upper screen will be associated with the text, and the text information that may be input is directly displayed, thereby avoiding the user repeatedly inputting the key input string, turning the page, etc., on the one hand, effectively reducing the user's key rate and improving Input efficiency, fast and convenient input, improve user experience, on the other hand, can reduce the response of the device to the user input key input string, page turning, etc., reducing the consumption of equipment system resources.
  • the user inputs the key input string "543486" in the input method, and the input method returns the candidate text information of the pinyin input corresponding to the key input string "543486” such as “marriage”, “result”, “fire” and “taken”.
  • the input method system interface suppose the user needs the "result” screen, but because of the wrong operation, select “married” on the screen, displayed in the application edit box, such as SMS edit box, chat window of instant messaging tool, etc. After that, I found that the input was wrong, so I retired twice and deleted the "married” that was already on the screen;
  • the frequency of occurrence of the data Py-AcB i such as the calculation result
  • the selection probability that the "result” is selected on the upper screen is P 1
  • the selection probability of the "fire” selected on the upper screen is P 2
  • the selection of the "taken over” is selected on the screen.
  • the user can select the desired candidate in the preferred way, and the result of the prior art is in the second selection. Since the input method system of the computer version brings the user the habit of the first candidate on the screen, the first candidate is selected more than the other. The candidate cost is much smaller; the other extreme case: suppose the user wants not the second choice "result", but the first candidate "received” on the second screen (relative to the first screen), then at In the prior art, the user also needs to click the page flip button once to select the desired text information on the second screen candidate. In this example, since “marriage” gives up the valuable position of the first screen, "receiving goods” can be first. The screen is displayed, thus avoiding a page flip key.
  • the user inputs the key input string "niyou” in the actual keyboard, and the input method returns the key input string “niyou” such as “you again”, “you have”, “you", “ni”, “he” and “greasy”.
  • the candidate text information of the corresponding pinyin input is displayed on the input method system interface, assuming that the user needs “you have” the upper screen, but the "you are” upper screen is incorrectly selected due to lack of concentration, etc., and is displayed in the application edit box. In the middle, I found that the input error, so backspace once, delete some of the "again” that has been on the screen;
  • B i to give "Yes”, “travel”, “from”, “Friends”, “right”, “mail” six B i; B i and, for each, B i are calculated in step 102 in the manner occurrence frequency is selected on the screen selection probability, i.e., i is calculated corresponding to the B quadruple of data Py-AcB i
  • the calculated "yes" selection probability screen on the selection of P 1 is calculated "yes” selection probability screen on the selection of P 1
  • "travel” selection probability screen on the selection of P 2 selected “by the” selection probability screen on the selection of P 3, "Friends”
  • the selection probability of the upper screen is P 4
  • the selection probability of the "right” selected upper screen is P 5
  • the selection probability of "mail” selected upper screen is P 6 , where P 1 >P 2 >P 3 >P 4 > P 5 >P 6 , sorted from high to low according to the selection probability of selecting the upper screen as “Yes”, “Swim”, “Yes”, “Friend”, “Right”, “Post”;
  • the user is using the instant messaging tool to chat with other users, inputting the key input string "96" in the Jiugong grid input keyboard, and the input method system returns key inputs such as "I”, " ⁇ ", “ ⁇ ”, “ ⁇ ” and "grip".
  • the candidate text information of the pinyin input corresponding to the string "96" is displayed on the input method system interface, and the current user selects "I” on the screen, which is displayed in the application editing box, and then the chat environment changes, so the backspace is deleted once. "I” that has been on the screen;
  • the selection probability of " ⁇ " being selected on the upper screen is P 4
  • the selection probability of "that” being selected on the upper screen is P 5
  • P 1 >P 2 >P 3 >P 4 >P 5 sorted according to the selection probability of the upper screen from high to low as “you”, “me”, “have”, “ ⁇ ”, “
  • P 5 is greater than the probability threshold p described in step 103 (that is, the selection probabilities of the five B i are greater than the preset probability threshold)
  • the five texts are displayed on the first screen according to the above sorting result.
  • the user can choose the preferred "you” upper screen.
  • User A and User B have a conversation at night.
  • User A intends to say “I am sleepy.”
  • I feel that it is not polite to say something, so I want to change to another saying "You “I am not sleepy,” and then deleted “I”, but User A feels that there is no need to face such a friend's sway, so I still want to say “I am sleepy”, that is, after deleting "I", enter “I” again. ".
  • FIG. 10 a structural block diagram of an embodiment of an input device for text information according to the present invention is shown, which may specifically include the following modules:
  • the text information determining module 1001 is configured to determine the first text information and the second text information when detecting that all or part of the on-screen text information is deleted; wherein the first text information may include all the on-screen text information Corresponding key input string, all on-screen text information and the number of backspaces used for deletion; the second text information may include text information that is re-screened after the first text information appears;
  • the selection probability calculation module 1002 is configured to calculate, according to the first text information, a selection probability that the second text information is selected on the upper screen;
  • the display module 1003 is configured to display the second text information in the specified location when the selection probability is greater than a preset probability threshold.
  • the selection probability calculation module 1002 may include the following modules:
  • a first searching module configured to search for a first historical appearance frequency of the first text information
  • a second search module configured to search for a second historical appearance frequency of the second text information after the occurrence of the first text information
  • a candidate determining module configured to determine, in the second text information, candidate second text information that is being displayed
  • the formula calculation module is configured to calculate a selection probability that the candidate second text information is selected on the screen by using the following formula:
  • Py-Ac is the first text information
  • Py is a key input string corresponding to all the upper screen text information
  • A is the all-on-screen text information
  • c is The number of backspaces
  • B i is the second text information
  • count(Py-Ac) is the first historical appearance frequency
  • count(Py-AcB i ) is the second historical appearance frequency
  • n is the The number of candidate second text information
  • i is a positive integer.
  • the apparatus may further include: a first recording module, a second recording module, a first statistic module, and a second statistic module.
  • the first recording module is configured to: when it is detected that all or part of the on-screen text information is deleted, the key input string corresponding to all the on-screen text information, all the on-screen text information, and the number of backspace records used for deletion For the first text message;
  • a second recording module configured to record the information of the text that is re-screened after the occurrence of the first text information is the second text information
  • a first statistic module configured to count a first historical appearance frequency of the first text information
  • the second statistic module is configured to count the second historical appearance frequency of the second text information after the occurrence of the first text information.
  • the candidate determining module may further include the following modules:
  • a frequency calculation sub-module configured to calculate an appearance frequency of the second text information; the appearance frequency includes a ratio of the second historical appearance frequency to the first historical appearance frequency;
  • an extraction submodule configured to extract the one or more second text information having the highest frequency of occurrence as the candidate second text information displayed in the first screen.
  • the display module 1003 may include the following modules:
  • a candidate presentation sub-module configured to display the candidate second text information in a specified location.
  • the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.
  • embodiments of the embodiments of the invention may be provided as a method, apparatus, or computer program product.
  • embodiments of the invention may be in the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware.
  • embodiments of the invention may take the form of a computer program product embodied on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • Embodiments of the invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing terminal device to produce a machine such that instructions are executed by a processor of a computer or other programmable data processing terminal device
  • Means are provided for implementing the functions specified in one or more of the flow or in one or more blocks of the flow chart.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the instruction device implements the functions specified in one or more blocks of the flowchart or in a flow or block of the flowchart.
  • FIG. 11 shows a structural block diagram of a terminal for executing an input method of text information according to an embodiment of the present invention.
  • the terminal conventionally includes a processor 1110 and a computer program product or computer readable medium in the form of a memory 1120.
  • the memory 1120 can be, for example, a flash memory, Electronic memory such as EEPROM (Electrically Erasable Programmable Read Only Memory), EPROM, hard disk or ROM.
  • Memory 1120 has a memory space 1130 for program code 1131 for performing any of the method steps described above.
  • the storage space 1130 for program code may include respective program codes 1131 for implementing various steps in the above methods, respectively.
  • the program code can be read from or written to one or more computer program products.
  • Such computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards or floppy disks.
  • Such a computer program product is typically a portable or fixed storage unit as described with reference to FIG.
  • the storage unit may have a storage section, a storage space, and the like arranged similarly to the storage 1120 in the terminal of FIG.
  • the program code can be compressed, for example, in an appropriate form.
  • the storage unit includes computer readable code 1131', i.e., code that can be read by, for example, a processor such as 1110, which when executed by a server causes the server to perform various steps in the methods described above.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • User Interface Of Digital Computer (AREA)
  • Input From Keyboards Or The Like (AREA)
  • Document Processing Apparatus (AREA)
  • Machine Translation (AREA)

Abstract

一种文本信息的输入方法及装置,所述的方法包括:当监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息(101);其中,所述第一文本信息包括全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数;所述第二文本信息包括所述第一文本信息出现之后所重新上屏的文本信息;依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率(102);当所述选择概率大于预设的概率阈值时,在指定的位置中展示所述第二文本信息(103)。该方法及装置能够有效地降低用户的按键率,提高输入效率,实现快速方便的输入,提高用户体验,可以减少设备系统资源的消耗。

Description

一种文本信息的输入方法和装置
本发明实施例要求在2014年7月11日提交中国专利局、申请号为201410332496.2、发明名称为“一种文本信息的输入方法和装置”的中国专利申请的优先权,其全部内容通过引用结合在本发明实施例中。
技术领域
本发明涉及输入法的技术领域,特别是涉及一种文本信息的输入方法和一种文本信息的输入装置。
背景技术
随着移动设备的广泛普及,用户经常使用移动设备处理各种事情,例如聊天、上网、发微博等等,随之用户经常需要使用输入法系统输入文字内容。
按照移动设备输入使用习惯,用户经常使用输入法九宫格键盘进行输入,其具有舒适大方的拇指按键,可以快速输入文字内容,并且降低了输入时的错误率。
用户在输入文字的时候,可能出现分神或者位置点击错误等原因造成误操作,选择了不是所需要的文字。例如,假设用户所需要的是“你敢”或者其他,但是由于误操作选择了“你好”进行输入上屏。
用户需要删除“你好”之后,再重新输入其他词,并且,用户可能还要通过翻页等方式才可以找到想要的文字,这不仅使得用户操作十分麻烦,降低输入效率,而且重新输入会增加移动设备资源的消耗。
发明内容
本发明实施例所要解决的技术问题是提供一种文本信息的输入方法,用以提高输入效率。
相应的,本发明实施例还提供了一种文本信息的输入装置,用以保证上述方法的实现及应用。
为了解决上述问题,本发明实施例公开了一种文本信息的输入方法,包括:
当监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息;其中,所述第一文本信息包括全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数;所述第二文本信息包括输入记录历史中所述第一文本信息出现之后所重新上屏的文本信息;
依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率;
当所述选择概率大于预设的概率阈值时,在指定的位置中展示所述第二文本信息。
优选地,所述依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率的步骤包括:
查找所述第一文本信息的第一历史出现频次;
查找出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次;
在所述第二文本信息中确定进行展示的候选第二文本信息;
采用以下公式计算所述候选第二文本信息被选择上屏的选择概率:
Figure PCTCN2015083756-appb-000001
其中,p为所述选择概率,Py-A-c为所述第一文本信息,Py为所述全部已上屏文本信息对应的按键输入串,A为所述全部已上屏文本信息,c为所述退格次数,Bi为所述第二文本信息,count(Py-A-c)为所述第一历史出现频次,count(Py-A-c-Bi)为所述第二历史出现频次,n为所述候选第二文本信息的数量,i为正整数。
优选地,所述第一历史出现频次和所述第二历史出现频次通过以下方式获得:
当监测到删除在先全部或部分已上屏文本信息时,将全部已上屏文本 信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数记录为第一文本信息;
记录在所述第一文本信息出现之后所重新上屏的文本信息为第二文本信息;
统计所述第一文本信息的第一历史出现频次;
统计出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次。
优选地,所述在所述第二文本信息中确定进行展示的候选第二文本信息的步骤包括:
计算所述第二文本信息的出现频率;所述出现频率包括所述第二历史出现频次与所述第一历史出现频次的比值;
提取所述出现频率最大的一个或多个第二文本信息作为在首屏中进行展示的候选第二文本信息。
优选地,所述在指定的位置中展示所述第二文本信息的步骤包括:
在指定的位置中展示所述候选第二文本信息。
本发明实施例还公开了一种文本信息的输入装置,包括:
文本信息确定模块,配置为在监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息;其中,所述第一文本信息包括全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数;所述第二文本信息包括输入记录历史中所述第一文本信息出现之后所重新上屏的文本信息;
选择概率计算模块,配置为依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率;
展示模块,配置为在所述选择概率大于预设的概率阈值时,在指定的位置中展示所述第二文本信息。
优选地,所述选择概率计算模块包括:
第一查找模块,配置为查找所述第一文本信息的第一历史出现频次;
第二查找模块,配置为查找出现所述第一文本信息之后再出现所述第 二文本信息的第二历史出现频次;
候选确定模块,配置为在所述第二文本信息中确定进行展示的候选第二文本信息;
公式计算模块,配置为采用以下公式计算所述候选第二文本信息被选择上屏的选择概率:
Figure PCTCN2015083756-appb-000002
其中,p为所述选择概率,Py-A-c为所述第一文本信息,Py为所述全部已上屏文本信息对应的按键输入串,A为所述全部已上屏文本信息,c为所述退格次数,Bi为所述第二文本信息,count(Py-A-c)为所述第一历史出现频次,count(Py-A-c-Bi)为所述第二历史出现频次,n为所述候选第二文本信息的数量,i为正整数。
优选地,所述装置还包括:
第一记录模块,配置为当监测到删除在先全部或部分已上屏文本信息时,将全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数记录为第一文本信息;
第二记录模块,配置为记录在所述第一文本信息出现之后所重新上屏的本文信息为第二文本信息;
第一统计模块,配置为统计所述第一文本信息的第一历史出现频次;
第二统计模块,配置为统计出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次。
优选地,所述候选确定模块包括:
出现频率计算子模块,配置为计算所述第二文本信息的出现频率;所述出现频率包括所述第二历史出现频次与所述第一历史出现频次的比值;
提取子模块,配置为提取所述出现频率最大的一个或多个第二文本信息作为在首屏中进行展示的候选第二文本信息。
优选地,所述展示模块包括:
候选展示子模块,配置为在指定的位置中展示所述候选第二文本信 息。
本发明实施例还公开了一种计算机程序,包括计算机可读代码,当所述计算机可读代码在终端上运行时,导致所述终端执行前述的文本信息的输入方法。
本发明实施例还公开了一种计算机可读介质,其中存储了前述的计算机程序。
与现有技术相比,本发明实施例包括以下优点:
本发明实施例在监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息,依据所述第一文本信息计算第二文本信息在首屏被选择上屏的选择概率,当选择概率大于预设的概率阈值时,在首屏中展示第二文本信息,充分利用了文本信息重码这一属性、用户上屏后删除文本信息这一反馈的信息,通过对用户退格后将会上屏的文本信息进行联想的方式,直接展示可能会输入的文本信息,避免用户重复输入按键输入串、翻页等操作,一方面,能够有效地降低用户的按键率,提高输入效率,实现快速方便的输入,提高用户体验,另一方面,可以减少设备对用户输入按键输入串、翻页等操作的响应,减少设备系统资源的消耗。
本发明实施例对退格行为后可能会输入的文本信息按概率从高到低进行排序,方便使得用户进行再次选择文本信息,提高了操作的简便性。
附图说明
图1是本发明的一种文本信息的输入方法实施例的步骤流程图;
图2是一种正常输入文本信息的流程示例图;
图3是一种在先还原场景的技术中输入文本信息的流程示例图;
图4是本发明的一种文本信息的输入流程示例图;
图5是一种正常输入文本信息的流程示例图;
图6是本发明的一种文本信息的输入流程示例图;
图7是一种正常输入文本信息的流程示例图;
图8是一种在先还原场景的技术中输入文本信息的流程示例图;
图9是本发明的一种文本信息的输入流程示例图;
图10是本发明的一种文本信息的输入装置实施例的结构框图;
图11是本发明的一种用于执行根据本发明实施例的文本信息的输入方法的终端的结构框图;以及,
图12是本发明的一种用于保持或者携带实现根据本发明实施例的文本信息的输入方法的程序代码的存储单元。
具体实施方式
为使本发明的上述目的、特征和优点能够更加明显易懂,下面结合附图和具体实施方式对本发明作进一步详细的说明。
用户在文本信息输入过程中时常由于输入错误或者错误选择文本信息上屏,而不得不退格删除误输部分而重新进行输入,以致降低输入效率且降低用户输入体验。
在实际应用中,经深入研究发现,用户在退格行为前后输入的文本信息是存在一定规律的,因此提出了本发明实施例的核心构思之一,利用用户退格行为进行分析并将用户需要重新上屏的文本信息提供给用户,从而有效利用用户的输入反馈信息提升用户输入效率,即本发明实施例可以提供一种文本信息在上屏后发生退格删除行为时的主动联想,并基于用户的退格行为分析,在用户删掉某些文本信息时,主动在首屏展示用户接下来很有可能会继续输入的文本信息。
当然,上述本发明实施例的核心构思之一,可以应用在文本信息已上屏的情形,也可以应用在文本信息未上屏的情形中,本发明实施例对此不加以限制。
参照图1,示出了本发明的一种文本信息的输入方法实施例的步骤流程图,可以包括如下步骤:
步骤101,当监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息;
其中,所述第一文本信息可以包括全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数;所述第二文本信息可以包括输入记录历史中所述第一文本信息出现之后所重新上屏的文本信 息;
需要说明的是,上屏可以为用户输入按键输入串获得候选文本信息后,点选候选文本信息的行为,退格可以为用户删除掉输入的文本信息的行为。该按键输入串可以为拼音、五笔、笔画、英文等输入方式对应的按键输入串,本发明实施例对此不加以限制;该文本信息可以为文字,也可以为数字,还可以为诸如“《”、“%”等符号,本发明实施例对此亦不加以限制。
在具体实现中,用户可以通过退格事件删除当前全部或部分已上屏文本信息,该退格事件可以通过退格键触发。退格键,也称回格键,可以用于使光标左移一格,同时删除光标左边位置上的字符。通常,退格键在键盘(包括现实的键盘和虚拟键盘)上的标识是“←”,英文字母是BackSpace。
应用本发明实施例,可以预先将输入记录历史中出现的第一文本信息Py-A-c(Py为所述全部已上屏文本信息对应的按键输入串,A为所述全部已上屏文本信息,c为所述退格次数)及其对应的第二文本信息Bi存储在数据库中。第一文本信息Py-A-c可以为三元组数据,第二文本信息可以为在该三元组数据出现的条件下所出现的事件,可以与该三元组数据组成四元组数据Py-A-c-Bi
在具体实现中,文本信息在输入过程中可能存在重码,即两个候选文本信息的按键输入串形式是一样的,例如,在键盘中的26个按键全拼形式一致(如“我”和“握”的拼音全拼的按键字符串都是“wo”)、在现实键盘中的26个按键简拼形式一致(如“主人”和“主任”的拼音简拼的按键字符串都是“zr”)、在九宫格键盘中的9个按键全拼形式一致(如“结果”和“结婚”的全拼的按键字符串都是“543486”)、在九宫格键盘中的9个按键简拼形式一致(如“主人”和“主任”的拼音简拼的按键字符串都是“97”)等等。
而全部已上屏文本信息对应的按键输入串Py可以用于确定重码。例如,用户在键盘中的26个按键的环境下输入的按键输入串为“woyou”, 上屏了“我有”,这个时候重码的文本信息有“我又”等等,但是不会出现“我走”;又例如,而在九宫格键盘中的9个按键的环境下输入的按键输入串为“96968”,重码的文本信息就有可能是“我走”,因此退格后联想的文本信息也可能不一样。
在实际应用中,输入法可能无法获知用户删除的文本信息,因此,全部已上屏文本信息A和退格次数c可以共同起作用,获知被用户删除的文本信息。例如:用户输入按键输入串“woyou”,选择“我有”上屏,如果c=1,即退格一次,则可以判断被删除的文本信息是“有”,如果c=2,即退格两次,即退格两次,则可以判断被删除的文本信息是“我有”。
当监测到删除当前全部或部分已上屏文本信息时,可以由该删除当前全部或部分已上屏文本信息确定当前全部已上屏文本信息对应的按键输入串Py,当前全部已上屏文本信息A,当前删除所用的退格次数c,获得三元组数据Py-A-c,即第一文本信息,再根据该三元组数据查找所有出现的四元组数据Py-A-c-Bi,即可确定第二文本信息Bi
其中,全部已上屏文本信息A和第二文本信息Bi可能相同,例如,很多用户会在删掉“我”之后会继续输入“我”。
步骤102,依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率;
需要说明的是,首屏可以为无需进行翻页及滑动等候选操作即可进行候选文本信息选择的输入法界面。
依据第一文本信息,从数据库中保存的历史输入记录中获取的第二文本信息可能有多个,则依据所述第一文本信息分别计算每个所述第二文本信息在输入记录历史中被选择上屏的选择概率。
在本发明的一种优选实施例中,步骤102可以包括如下子步骤:
子步骤S11,查找所述第一文本信息的第一历史出现频次;
子步骤S12,查找出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次;
在具体实现中,第一历史出现频次可以为出现第一文本信息这一事件 在记录历史中所出现的次数,也即第一历史频次为在输入所述全部已上屏文本信息对应的按键输入串,并选择所述全部已上屏文本信息上屏后,执行所述退格次数次退格的行为在记录历史中所出现的频次;第二历史出现频次可以为在出现第一文本信息之后再出现第二文本信息这一事件在记录历史中所出现的次数,也即第二历史频次为在输入所述全部已上屏文本信息对应的按键输入串,选择所述全部已上屏文本信息上屏,并执行所述退格次数次的退格后,又选择所述重新上屏的文本信息上屏的行为在记录历史中所出现的频次。
在本发明实施例的一种优选示例中,所述第一历史出现频次和所述第二历史出现频次可以通过以下方式获得:
子步骤S21,当监测到删除在先全部或部分已上屏文本信息时,将全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数记录为第一文本信息;
子步骤S22,记录在所述第一文本信息出现之后所重新上屏的第二文本信息;
子步骤S23,统计所述第一文本信息的第一历史出现频次;
子步骤S24,统计出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次。
进一步的,将上述第一文本信息及对应的第一历史出现频次、第二文本信息及对应的第二历史出现频次作为输入记录历史保存在数据库中。以便后续执行子步骤S11和子步骤S12的查找操作。
优选地,每个用户的输入文本信息的习惯都不尽相同,因此为了满足用户个性化的需求,进一步提高第二文本信息命中用户需求的概率,第一历史出现频次和第二历史出现频次可以为记录的当前用户输入的第一文本信息的第一历史出现频次和当前用户在输入第一文本信息之后再重新上屏的第二文本信息的第二历史出现频次。
当然,同一用户在不同情形下输入的第一文本信息和第二文本信息可能存在不均匀的情况,例如聊天类的文本信息较大,但是购物类的文本信 息可能较少,容易出现数据稀疏的问题,因此为了避免数据稀疏的情况,本申请中可以基于全体用户输入的第一文本信息的情况确定所述第一历史出现频次,也可以基于全体用户在输入第一文本信息之后再重新上屏的第二文本信息的情况来确定所述第二历史出现频次。
例如,第一历史出现频次可以为记录的全体用户输入的第一文本信息的历史出现频次。第二历史出现频次可以为记录的全体用户在输入第一文本信息之后再重新上屏第二文本信息的历史出现频次。
再例如,第一历史出现频次也可以为记录的全体用户输入的第一文本信息的历史出现频次,与当前用户输入的第一文本信息的历史出现频次,按照一定比例混合而成。第二历史出现频次也可以为全体用户在输入第一文本信息之后再重新上屏第二文本信息的历史出现频次,与当前用户输入第一文本信息之后再重新上屏第二文本信息的历史出现频次,按照一定比例混合而成的。
本发明实施例对此不加以限制。通过上述方式,不再单独考虑当前用户输入的第一文本信息和第二文本信息的情况,而是考虑到了全体用户的情况,全体用户整体输入的信息更加均匀,因此能够避免当前用户输入的信息不均匀而导致得到的第一历史出现频次和第二历史出现频次不准确的情况。
在具体实现中,若用户点击了退格键删除了已上屏的文本信息,则累加退格次数c,至用户下一次动作时,若c不为0,则进入退格联想的训练过程,即统计第一文本信息的第一历史出现频次count(Py-A-c)和出现所述第一文本信息之后再出现第二文本信息的第二历史出现频次count(Py-A-c-Bi),否则认为用户正常输入文本信息,不进行退格联想的训练过程。
子步骤S13,在所述第二文本信息中确定进行展示的候选第二文本信息;
在实际应用中,由于用于展示的位置有限,因此需要确定进行展示的候选第二文本信息。
在本发明实施例的一种优选示例中,子步骤S13进一步可以包括如下 子步骤:
子步骤S131,计算所述第二文本信息的出现频率;所述出现频率可以包括所述第二历史出现频次与所述第一历史出现频次的比值;
子步骤S132,提取所述出现频率最大的一个或多个第二文本信息作为在首屏中进行展示的候选第二文本信息。
在具体实现中,可以根据四元组数据的出现频率由高到低进行排序,然后确定首屏可以容纳的第二文本信息的个数,例如n个,n为正整数,再在该排序中提取前n个第二文本信息作为在首屏中进行展示的候选第二文本信息。
本发明实施例对退格行为后可能会上屏的文本信息按概率从高到低进行排序,方便使得用户进行再次选择文本信息,提高了操作的简便性。
子步骤S14,采用以下公式计算所述候选第二文本信息被选择上屏的选择概率:
Figure PCTCN2015083756-appb-000003
其中,p为所述选择概率,Py-A-c为所述第一文本信息,Py为所述全部已上屏文本信息对应的按键输入串,A为所述全部已上屏文本信息,c为所述退格次数,Bi为所述第二文本信息,count(Py-A-c)为所述第一历史出现频次,count(Py-A-c-Bi)为所述第二历史出现频次,n为所述候选第二文本信息的数量,i为正整数。
步骤103,当所述选择概率大于预设的概率阈值时,在指定的位置展示所述第二文本信息。
在具体实现中,指定的位置可以为首屏,也可以是联想区域(例如在首屏附近生成气泡等形状的界面),还可以是设备当前界面的任意区域,本发明实施例对此不加以限制。
在本发明的一种优选实施例中,步骤103可以包括如下子步骤:
子步骤S31,在指定的位置展示所述候选第二文本信息。
若选择概率p大于概率阈值,表示超过一定的概率能在首屏展示的第 二文本信息中命中用户接下来将要上屏的文本信息,则可以在首屏展示候选第二文本信息;若选择概率p小于或等于概率阈值,表示能在首屏展示的第二文本信息中命中用户接下来将要上屏的文本信息概率过低,则可以放弃在首屏展示候选第二文本信息。
本发明实施例在监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息,依据所述第一文本信息计算第二文本信息在首屏被选择上屏的选择概率,当选择概率大于预设的概率阈值时,在首屏中展示第二文本信息,充分利用了文本信息重码这一属性、用户上屏后删除文本信息这一反馈的信息,通过对用户退格后将会上屏的文本信息进行联想的方式,直接展示可能会输入的文本信息,避免用户重复输入按键输入串、翻页等操作,一方面,能够有效地降低用户的按键率,提高输入效率,实现快速方便的输入,提高用户体验,另一方面,可以减少设备对用户输入按键输入串、翻页等操作的响应,减少设备系统资源的消耗。
以下通过几种应用场景的示例对本发明实施例作进一步说明:
应用场景一
用户在输入法九宫格键盘输入按键输入串“543486”,输入法系统返回“结婚”、“结果”、“烈火”和“接过”等按键输入串“543486”对应的拼音输入的候选文本信息,显示在输入法系统界面上,假设用户需要“结果”上屏,但是因为误操作选择“结婚”上屏,显示在应用程序编辑框中,例如短信编辑框、即时通讯工具的聊天窗口等等,之后发现输入错误,因此退格两次,删除已上屏的“结婚”;
在正常输入的流程中,如图2所示,接着用户重输按键输入串“543486”,获得与在先的结果一样的候选文本信息,“结婚”、“结果”、“烈火”和“接过”,即首选“结婚”、次选“结果”,用户可以选择次选“结果”上屏;
应用在先还原场景的技术,如图3所示,用户无需再次输入拼音串“543486”,输入法系统自动展现原来的候选文本信息,“结婚”、“结 果”、“烈火”和“接过”,依然首选“结婚”、次选“结果”,用户可以选择次选“结果”上屏;
应用本发明实施例,如图4所示,确定当前全部已上屏文本信息对应的按键输入串Py为“543486”,当前全部已上屏文本信息A为“结婚”,当前删除所用的退格次数c为“2”,确定被删除的文本信息为“结婚”,则三元组数据Py-A-c为“543486-结婚-2”,首先根据该三元组数据从步骤101中所述的数据库中查找所有出现的四元组数据Py-A-c-Bi,即查询在三元组数据“543486-结婚-2”出现时所重新上屏的文本信息Bi,得到“结果”、“烈火”、“接过”和“接货”这4个Bi;然后针对每个Bi,分别按照步骤102的方式计算该Bi被选择上屏的选择概率,也即计算该Bi对应的四元组数据Py-A-c-Bi的出现频率,如计算得到“结果”被选择上屏的选择概率为P1,“烈火”被选择上屏的选择概率为P2,“接过”被选择上屏的选择概率为P3,“接货”被选择上屏的选择概率为P4,其中P1>P2>P3>P4,根据选择上屏的选择概率由高到低进行排序为“结果”、“烈火”、“接过”、“接货”;最后如果P4大于步骤103中所述的概率阈值p(也即这4个Bi的选择概率均大于预设的概率阈值),则按照上述排序结果在首屏展示这4个文本,其中“结果”位于首选(如图4所示),用户可以选择首选“结果”上屏。
就本例而言,本示例相对于在先还原场景的技术(以下简称“在先技术”)的优点在于:
1、用户可以在首选选择想要的候选,而在先技术的结果在次选,由于电脑版本等的输入法系统带给了用户空格上屏首个候选的习惯,选择首个候选会比其他候选代价小很多;另外一个极端的情况:假设用户想要的不是次选“结果”,而是位于第二屏(相对于首屏而言)的第一个候选“接货”,则在在先技术中,用户还需要点击一次翻页键,在第二屏候选选择想要的文本信息,而在本示例,由于“结婚”让出了首屏的宝贵位置,“接货”得以在首屏展现,从而避免了一次翻页键。
2、用户在输入不同的内容时,出错的概率不一样,很有可能退格后的 输入分布和正常输入时的分布有着很大的差别,所以用正常输入的输入分布来模拟退格后的输入分布会有较大的误差,而在先技术正是拿正常输入的上屏分布来对退格后上屏的文本信息进行预测,用户误将文本信息上屏之后,点击了删除键,其实已经给了输入法系统一些反馈信息,如果真的是误上屏了,那么刚刚上屏的文本信息是不会再次被输入的,因而在先技术不能最大限度地利用首屏有限的展示空间,这种现象在文本信息比较接近(例如形近字、音近字等)的时候会尤为明显。
应用场景二
用户在现实的键盘中输入按键输入串“niyou”,输入法系统返回“你又”、“你有”、“你”、“尼”、“呢”和“腻”等按键输入串“niyou”对应的拼音输入的候选文本信息,显示在输入法系统界面上,假设用户需求“你有”上屏,但是由于精神不集中等原因错误选择了“你又”上屏,显示在应用程序编辑框中,之后发现输入错误,因此退格一次,删除部分已上屏的“又”;
在正常输入的流程中,如图5所示,接着用户重输按键输入串“you”,获得候选文本信息,“有”、“又”、“油”、“you”、“游”和“友”,用户可以选择“有”上屏;
由于删除的是部分已上屏的文本信息,因此无法触发在先还原场景的技术;
应用本发明实施例,如图6所示,确定当前全部已上屏文本信息对应的按键输入串Py为“niyou”,当前全部已上屏文本信息A为“你又”,当前删除所用的退格次数c为“1”,确定被删除的文本信息为“又”,虽然与“又”重码的文本信息很多,例如“由”、“优”、“友”等等,但是与上文“你”的存在可以有效地移除可信度较低的文本信息,例如“优”、“友”等等,则三元组数据Py-A-c为“niyou-你又-1”,首先根据该三元组数据从步骤101中所述的数据库中查找所有出现的四元组数据Py-A-c-Bi,即查询在三元组数据“niyou-你又-1”出现时所重新上屏的文 本信息Bi,得到“有”、“游”、“由”、“友”、“右”、“邮”这6个Bi;然后针对每个Bi,分别按照步骤102的方式计算该Bi被选择上屏的选择概率,也即计算该Bi对应的四元组数据Py-A-c-Bi的出现频率,如计算得到“有”被选择上屏的选择概率为P1,“游”被选择上屏的选择概率为P2,“由”被选择上屏的选择概率为P3,“友”被选择上屏的选择概率为P4,“右”被选择上屏的选择概率为P5,“邮”被选择上屏的选择概率为P6,其中P1>P2>P3>P4>P5>P6,根据选择上屏的选择概率由高到低进行排序为“有”、“游”、“由”、“友”、“右”、“邮”;最后如果P6大于步骤103中所述的概率阈值p(也即这6个Bi的选择概率均大于预设的概率阈值),则按照上述排序结果在首屏展示这6个文本,其中“有”位于首选(如图6所示),用户可以选择“有”上屏。
应用场景三
用户在使用即时通讯工具与其他用户聊天,在九宫格输入键盘中输入按键输入串“96”,输入法系统返回“我”、“哟”、“喔”、“窝”和“握”等按键输入串“96”对应的拼音输入的候选文本信息,显示在输入法系统界面上,当前用户选择“我”上屏,显示在应用程序编辑框中,之后聊天环境发生变化,因此退格一次,删除已上屏的“我”;
在正常输入的流程中,如图7所示,接着用户重输按键输入串“64”,获得候选文本信息,“你”、“拟”、“泥”、“尼”和“腻”,用户可以选择“你”上屏;
应用在先还原场景的技术,如图8所示,输入法系统再次展现原来的候选文本信息,“我”、“哟”、“喔”、“窝”和“握”,并没有用户所需的“你”;
应用本发明实施例,如图9所示,确定当前全部已上屏文本信息对应的按键输入串Py为“wo”,当前全部已上屏文本信息A为“我”,当前删除所用的退格次数c为“1”,确定被删除的文本信息为“我”,则三元组数据Py-A-c为“wo-我-1”,首先根据该三元组数据从步骤101中所述的数 据库中查找所有出现的四元组数据Py-A-c-Bi,即查询在三元组数据“wo-我-1”出现时所输入的文本信息Bi,得到“你”、“我”、“有”、“哟”和“那”这5个Bi;然后针对每个Bi,分别按照步骤102的方式计算该Bi被选择上屏的选择概率,也即计算该Bi对应的四元组数据Py-A-c-Bi的出现频率,如计算得到“你”被选择上屏的选择概率为P1,“我”被选择上屏的选择概率为P2,“有”被选择上屏的选择概率为P3,“哟”被选择上屏的选择概率为P4,“那”被选择上屏的选择概率为P5,其中P1>P2>P3>P4>P5,根据选择上屏的选择概率由高到低进行排序为“你”、“我”、“有”、“哟”、“那”;最后如果P5大于步骤103中所述的概率阈值p(也即这5个Bi的选择概率均大于预设的概率阈值),则按照上述排序结果在首屏展示这5个文本,其中“你”位于首选(如图9所示),用户可以选择首选“你”上屏。
用户在即时通讯工具中的聊天窗口进行会话经常发生思维转换,在“我”和“你”等词之间发生思维转换。
例如,用户A与用户B在夜晚进行会话,用户A打算想说“我困了”,在输入完“我”之后,觉得想说的话不太礼貌,因此想换成另外一种说法“你困不困”,即需要删除“我”之后,再输入“你”。
又例如,用户A与用户B在夜晚进行会话,用户A打算想说“我困了”,在输入完“我”之后,觉得想说的话不太礼貌,因此想换成另外一种说法“你困不困”,随即删除了“我”,但是用户A又觉得没必要面对朋友说得这么婉转,因此还是想说“我困了”,即在删除了“我”之后,再次输入“我”。
而事实上,在即时通讯工具之类的应用里,集中了绝大部分的用户输入行为,用户揣测语言、欲言又止的情况经常发生,涉及到“我”和“你”这样的极高频文本信息发生思维转换的情况很多,而且跳转的文本信息也相对较为分散,不完全是重码部分的文本信息。
需要说明的是,对于方法实施例,为了简单描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本发明实施例并不受所 描述的动作顺序的限制,因为依据本发明实施例,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作并不一定是本发明实施例所必须的。
参照图10,示出了本发明一种文本信息的输入装置实施例的结构框图,具体可以包括如下模块:
文本信息确定模块1001,配置为在监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息;其中,所述第一文本信息可以包括全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数;所述第二文本信息可以包括所述第一文本信息出现之后所重新上屏的文本信息;
选择概率计算模块1002,配置为依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率;
展示模块1003,配置为在所述选择概率大于预设的概率阈值时,在指定的位置中展示所述第二文本信息。
在本发明的一种优选实施例中,所述选择概率计算模块1002可以包括如下模块:
第一查找模块,配置为查找所述第一文本信息的第一历史出现频次;
第二查找模块,配置为查找出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次;
候选确定模块,配置为在所述第二文本信息中确定在进行展示的候选第二文本信息;
公式计算模块,配置为采用以下公式计算所述候选第二文本信息被选择上屏的选择概率:
Figure PCTCN2015083756-appb-000004
其中,p为所述选择概率,Py-A-c为所述第一文本信息,Py为所述全 部已上屏文本信息对应的按键输入串,A为所述全部已上屏文本信息,c为所述退格次数,Bi为所述第二文本信息,count(Py-A-c)为所述第一历史出现频次,count(Py-A-c-Bi)为所述第二历史出现频次,n为所述候选第二文本信息的数量,i为正整数。
在本发明实施例的一种优选示例中,所述装置还可以包括:第一记录模块、第二记录模块、第一统计模块和第二统计模块。
第一记录模块,配置为当监测到删除在先全部或部分已上屏文本信息时,将全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数记录为第一文本信息;
第二记录模块,配置为记录在所述第一文本信息出现之后所重新上屏的本文信息为第二文本信息;
第一统计模块,配置为统计所述第一文本信息的第一历史出现频次;
第二统计模块,配置为统计出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次。
在本发明实施例的一种优选示例中,所述候选确定模块进一步可以包括如下模块:
出现频率计算子模块,配置为计算所述第二文本信息的出现频率;所述出现频率包括所述第二历史出现频次与所述第一历史出现频次的比值;
提取子模块,配置为提取所述出现频率最大的一个或多个第二文本信息作为在首屏中进行展示的候选第二文本信息。
在本发明的一种优选实施例中,所述展示模块1003可以包括如下模块:
候选展示子模块,配置为在指定的位置中展示所述候选第二文本信息。
对于装置实施例而言,由于其与方法实施例基本相似,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
本说明书中的各个实施例均采用递进的方式描述,每个实施例重点说明 的都是与其他实施例的不同之处,各个实施例之间相同相似的部分互相参见即可。
本领域内的技术人员应明白,本发明实施例的实施例可提供为方法、装置、或计算机程序产品。因此,本发明实施例可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本发明实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明实施例是参照根据本发明实施例的方法、终端设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理终端设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理终端设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理终端设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理终端设备上,使得在计算机或其他可编程终端设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程终端设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
例如,图11示出了用于执行根据本发明实施例的文本信息的输入方法的终端的结构框图。该终端传统上包括处理器1110和以存储器1120形式的计算机程序产品或者计算机可读介质。存储器1120可以是诸如闪存、 EEPROM(电可擦除可编程只读存储器)、EPROM、硬盘或者ROM之类的电子存储器。存储器1120具有用于执行上述方法中的任何方法步骤的程序代码1131的存储空间1130。例如,用于程序代码的存储空间1130可以包括分别用于实现上面的方法中的各种步骤的各个程序代码1131。这些程序代码可以从一个或者多个计算机程序产品中读出或者写入到这一个或者多个计算机程序产品中。这些计算机程序产品包括诸如硬盘,紧致盘(CD)、存储卡或者软盘之类的程序代码载体。这样的计算机程序产品通常为如参考图12所述的便携式或者固定存储单元。该存储单元可以具有与图11的终端中的存储器1120类似布置的存储段、存储空间等。程序代码可以例如以适当形式进行压缩。通常,存储单元包括计算机可读代码1131’,即可以由例如诸如1110之类的处理器读取的代码,这些代码当由服务器运行时,导致该服务器执行上面所描述的方法中的各个步骤。
尽管已描述了本发明实施例的优选实施例,但本领域内的技术人员一旦得知了基本创造性概念,则可对这些实施例做出另外的变更和修改。所以,所附权利要求意欲解释为包括优选实施例以及落入本发明实施例范围的所有变更和修改。
最后,还需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者终端设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者终端设备所固有的要素。在没有更多限制的情况下,由语句“包括一个......”限定的要素,并不排除在包括所述要素的过程、方法、物品或者终端设备中还存在另外的相同要素。
以上对本发明所提供的一种文本信息的输入方法和一种文本信息的输入装置,进行了详细介绍,本文中应用了具体个例对本发明的原理及实施 方式进行了阐述,以上实施例的说明只是用于帮助理解本发明的方法及其核心思想;同时,对于本领域的一般技术人员,依据本发明的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本发明的限制。

Claims (12)

  1. 一种文本信息的输入方法,包括:
    当监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息;其中,所述第一文本信息包括全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数;所述第二文本信息包括输入记录历史中所述第一文本信息出现之后所重新上屏的文本信息;
    依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率;
    当所述选择概率大于预设的概率阈值时,在指定的位置中展示所述第二文本信息。
  2. 根据权利要求1所述的方法,其中,所述依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率的步骤包括:
    查找所述第一文本信息的第一历史出现频次;
    查找出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次;
    在所述第二文本信息中确定进行展示的候选第二文本信息;
    采用以下公式计算所述候选第二文本信息被选择上屏的选择概率:
    Figure PCTCN2015083756-appb-100001
    其中,p为所述选择概率,Py-A-c为所述第一文本信息,Py为所述全部已上屏文本信息对应的按键输入串,A为所述全部已上屏文本信息,c为所述退格次数,Bi为所述第二文本信息,count(Py-A-c)为所述第一历史出现频次,count(Py-A-c-Bi)为所述第二历史出现频次,n为所述候选第二文本信息的数量,i为正整数。
  3. 根据权利要求2所述的方法,其中,所述第一历史出现频次和所述第二历史出现频次通过以下方式获得:
    当监测到删除在先全部或部分已上屏文本信息时,将全部已上屏文本 信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数记录为第一文本信息;
    记录在所述第一文本信息出现之后所重新上屏的文本信息为第二文本信息;
    统计所述第一文本信息的第一历史出现频次;
    统计出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次。
  4. 根据权利要求2所述的方法,其中,所述在所述第二文本信息中确定进行展示的候选第二文本信息的步骤包括:
    计算所述第二文本信息的出现频率;所述出现频率包括所述第二历史出现频次与所述第一历史出现频次的比值;
    提取所述出现频率最大的一个或多个第二文本信息作为在首屏中进行展示的候选第二文本信息。
  5. 根据权利要求2所述的方法,其中,所述在指定的位置中展示所述第二文本信息的步骤包括:
    在指定的位置中展示所述候选第二文本信息。
  6. 一种文本信息的输入装置,包括:
    文本信息确定模块,配置为在监测到删除当前全部或部分已上屏文本信息时,确定第一文本信息和第二文本信息;其中,所述第一文本信息包括全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数;所述第二文本信息包括输入记录历史中所述第一文本信息出现之后所重新上屏的文本信息;
    选择概率计算模块,配置为依据所述第一文本信息计算所述第二文本信息被选择上屏的选择概率;
    展示模块,配置为在所述选择概率大于预设的概率阈值时,在指定的位置中展示所述第二文本信息。
  7. 根据权利要求6所述的装置,其中,所述选择概率计算模块包括:
    第一查找模块,配置为查找所述第一文本信息的第一历史出现频次;
    第二查找模块,配置为查找出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次;
    候选确定模块,配置为在所述第二文本信息中确定进行展示的候选第二文本信息;
    公式计算模块,配置为采用以下公式计算所述候选第二文本信息被选择上屏的选择概率:
    Figure PCTCN2015083756-appb-100002
    其中,p为所述选择概率,Py-A-c为所述第一文本信息,Py为所述全部已上屏文本信息对应的按键输入串,A为所述全部已上屏文本信息,c为所述退格次数,Bi为所述第二文本信息,count(Py-A-c)为所述第一历史出现频次,count(Py-A-c-Bi)为所述第二历史出现频次,n为所述候选第二文本信息的数量,i为正整数。
  8. 根据权利要求7所述的装置,其中,所述装置还包括:
    第一记录模块,配置为当监测到删除在先全部或部分已上屏文本信息时,将全部已上屏文本信息对应的按键输入串、全部已上屏文本信息和删除所用的退格次数记录为第一文本信息;
    第二记录模块,配置为记录在所述第一文本信息出现之后所重新上屏的本文信息为第二文本信息;
    第一统计模块,配置为统计所述第一文本信息的第一历史出现频次;
    第二统计模块,配置为统计出现所述第一文本信息之后再出现所述第二文本信息的第二历史出现频次。
  9. 根据权利要求7所述的装置,其中,所述候选确定模块包括:
    出现频率计算子模块,配置为计算所述第二文本信息的出现频率;所述出现频率包括所述第二历史出现频次与所述第一历史出现频次的比值;
    提取子模块,配置为提取所述出现频率最大的一个或多个第二文本信息作为在首屏中进行展示的候选第二文本信息。
  10. 根据权利要求9所述的装置,其中,所述展示模块包括:
    候选展示子模块,配置为在指定的位置中展示所述候选第二文本信息。
  11. 一种计算机程序,包括计算机可读代码,当所述计算机可读代码在终端上运行时,导致所述终端执行根据权利要求1-5中的任一个所述的文本信息的输入方法。
  12. 一种计算机可读介质,其中存储了如权利要求11所述的计算机程序。
PCT/CN2015/083756 2014-07-11 2015-07-10 一种文本信息的输入方法和装置 WO2016004893A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2016564625A JP6667452B2 (ja) 2014-07-11 2015-07-10 テキスト情報の入力方法及び装置
KR1020167031676A KR102327418B1 (ko) 2014-07-11 2015-07-10 텍스트 정보의 입력 방법 및 장치

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410332496.2 2014-07-11
CN201410332496.2A CN105334952B (zh) 2014-07-11 2014-07-11 一种文本信息的输入方法和装置

Publications (1)

Publication Number Publication Date
WO2016004893A1 true WO2016004893A1 (zh) 2016-01-14

Family

ID=55063602

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/083756 WO2016004893A1 (zh) 2014-07-11 2015-07-10 一种文本信息的输入方法和装置

Country Status (4)

Country Link
JP (1) JP6667452B2 (zh)
KR (1) KR102327418B1 (zh)
CN (1) CN105334952B (zh)
WO (1) WO2016004893A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111221603A (zh) * 2019-10-30 2020-06-02 深圳前海微众银行股份有限公司 一种基于输入框的信息输入方法及装置

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112306253A (zh) * 2019-07-30 2021-02-02 北京搜狗科技发展有限公司 一种确定输入对象的方法、装置及设备
CN112445349A (zh) * 2019-09-05 2021-03-05 北京搜狗科技发展有限公司 一种输入方法、装置和电子设备
CN113534973B (zh) * 2020-04-16 2024-08-02 北京搜狗科技发展有限公司 输入方法、装置和用于输入的装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101241514A (zh) * 2008-03-21 2008-08-13 北京搜狗科技发展有限公司 一种生成纠错数据库的方法、自动纠错的方法和系统
CN101350004A (zh) * 2008-09-11 2009-01-21 北京搜狗科技发展有限公司 形成个性化纠错模型的方法及个性化纠错的输入法系统
CN102156551A (zh) * 2011-03-30 2011-08-17 北京搜狗科技发展有限公司 一种字词输入的纠错方法及系统
CN103049458A (zh) * 2011-10-17 2013-04-17 北京搜狗科技发展有限公司 一种修正用户词库的方法和系统
EP2653982A1 (en) * 2012-04-20 2013-10-23 King Abdulaziz City for Science & Technology (KACST) Method and system for statistical misspelling correction
CN104461042A (zh) * 2013-09-16 2015-03-25 百度在线网络技术(北京)有限公司 基于退回键自动进行纠错的日文输入方法及系统

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3003605B2 (ja) * 1996-12-25 2000-01-31 日本電気株式会社 単語修正装置および単語修正方法
JP2003131776A (ja) * 2001-10-29 2003-05-09 Sony Corp 情報処理装置、文字列入力装置、および確定履歴制御方法
KR100736561B1 (ko) * 2005-12-28 2007-07-09 엘지전자 주식회사 오타수정장치와 그 방법 및 이동통신단말기
KR101465770B1 (ko) * 2007-06-25 2014-11-27 구글 인코포레이티드 단어 확률 결정
CN100555203C (zh) * 2008-09-26 2009-10-28 腾讯科技(深圳)有限公司 修正已输入文字的系统及方法
CN100559332C (zh) * 2008-10-17 2009-11-11 腾讯科技(深圳)有限公司 恢复已删文字的系统及方法
CN101989128B (zh) * 2009-08-03 2014-07-23 北京搜狗科技发展有限公司 一种在固定位置展现候选词的方法及装置
CN102193645B (zh) * 2010-03-16 2017-07-18 深圳市世纪光速信息技术有限公司 文字输入方法和文字输入系统
JP5931627B2 (ja) * 2012-07-23 2016-06-08 京セラ株式会社 携帯端末装置、プログラムおよび入力訂正方法
JP2014191355A (ja) * 2013-03-26 2014-10-06 Oki Data Corp 文字入力装置及び文字入力方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101241514A (zh) * 2008-03-21 2008-08-13 北京搜狗科技发展有限公司 一种生成纠错数据库的方法、自动纠错的方法和系统
CN101350004A (zh) * 2008-09-11 2009-01-21 北京搜狗科技发展有限公司 形成个性化纠错模型的方法及个性化纠错的输入法系统
CN102156551A (zh) * 2011-03-30 2011-08-17 北京搜狗科技发展有限公司 一种字词输入的纠错方法及系统
CN103049458A (zh) * 2011-10-17 2013-04-17 北京搜狗科技发展有限公司 一种修正用户词库的方法和系统
EP2653982A1 (en) * 2012-04-20 2013-10-23 King Abdulaziz City for Science & Technology (KACST) Method and system for statistical misspelling correction
CN104461042A (zh) * 2013-09-16 2015-03-25 百度在线网络技术(北京)有限公司 基于退回键自动进行纠错的日文输入方法及系统

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111221603A (zh) * 2019-10-30 2020-06-02 深圳前海微众银行股份有限公司 一种基于输入框的信息输入方法及装置

Also Published As

Publication number Publication date
JP2017528777A (ja) 2017-09-28
JP6667452B2 (ja) 2020-03-18
CN105334952A (zh) 2016-02-17
CN105334952B (zh) 2018-12-18
KR20170028875A (ko) 2017-03-14
KR102327418B1 (ko) 2021-11-17

Similar Documents

Publication Publication Date Title
JP6431119B2 (ja) 携帯式端末設備における摺接操作による入力補助制御のシステム及び方法
US10156981B2 (en) User-centric soft keyboard predictive technologies
CN108370347B (zh) 对传入通信的预测响应方法和系统
US9424668B1 (en) Session-based character recognition for document reconstruction
JP5706137B2 (ja) 複数の投稿(データのグループ)を複数の軸に沿ってリアルタイムにコンピュータ画面に表示する方法およびコンピュータ・プログラム
US10417492B2 (en) Conversion of static images into interactive maps
US20150234938A1 (en) Method and electronic terminal for searching for contact in directory
EP3262529A1 (en) Topically aware word suggestions
US20140160032A1 (en) Swipe Stroke Input and Continuous Handwriting
WO2016004893A1 (zh) 一种文本信息的输入方法和装置
CN104199607A (zh) 一种基于输入法的候选选择方法和装置
WO2016095645A1 (zh) 笔画输入方法、装置和系统
CN105094603B (zh) 一种关联输入的方法与装置
KR20150023151A (ko) 전자 장치 및 이의 어플리케이션 실행 방법
WO2023061276A1 (zh) 数据推荐方法、装置、电子设备及存储介质
CN105589570B (zh) 一种处理输入错误的方法和装置
WO2016131425A1 (zh) 滑行输入方法及装置
CN104240107B (zh) 社群数据筛选系统及其方法
CN104076945A (zh) 用于在终端中显示输入键盘的装置和方法
WO2016184353A1 (zh) 一种在表单中检索定位信息的方法和装置
CN113805708A (zh) 信息显示方法、装置、电子设备和存储介质
CN113010072A (zh) 搜索方法、装置、电子设备及可读存储介质
CN111367424A (zh) 信息候选项智能化调整方法、装置及计算机可读存储介质
CN113126781A (zh) 信息显示方法、装置、设备和存储介质
CN114356118A (zh) 字符输入方法、装置、电子设备及介质

Legal Events

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

Ref document number: 15819220

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2016564625

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20167031676

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15819220

Country of ref document: EP

Kind code of ref document: A1