CN108182000A - Keyboard input detection method and device, storage medium and electronic equipment - Google Patents
Keyboard input detection method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN108182000A CN108182000A CN201711418534.6A CN201711418534A CN108182000A CN 108182000 A CN108182000 A CN 108182000A CN 201711418534 A CN201711418534 A CN 201711418534A CN 108182000 A CN108182000 A CN 108182000A
- Authority
- CN
- China
- Prior art keywords
- input
- score
- abnormal
- behavior
- keyboard input
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
- G06F3/023—Arrangements 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
- G06F3/0233—Character input methods
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)
- Input From Keyboards Or The Like (AREA)
Abstract
The disclosure provides a keyboard input detection method and device, a storage medium and an electronic device. The method comprises the following steps: acquiring a keyboard input behavior of a user, wherein the keyboard input behavior comprises an input speed and/or an input content; obtaining an abnormal score of the keyboard input behavior based on the historical input habit of the user; and if the abnormal score exceeds a preset score threshold value, judging that the keyboard input behavior belongs to abnormal input. According to the scheme, abnormal keyboard input detection can be realized, and special limits on input contents, character lengths, used languages and the like are not required.
Description
Technical Field
The present disclosure relates to the field of human-computer interaction technologies, and in particular, to a keyboard input detection method and apparatus, a storage medium, and an electronic device.
Background
Currently, keyboard input is widely used in various devices as the most common man-machine interaction mode. When the keyboard is in a use state, some contents may be input by mistake due to the triggering of an emergency, for example, a mobile phone placed in a pocket touches the keyboard by mistake under the condition that the screen is not locked; the mobile phone placed on the dining table is splashed with waterlogging under the condition of not locking the screen, and the virtual keyboard is triggered by mistake; a child or pet may scramble a keyboard, enter some meaningless strings, and so on.
Generally, as long as the device monitors input, the device records input content and sends the recorded content when a sending key is triggered by mistake, which inevitably reduces the use experience of the user. In response, the prior art provides the following scheme for recognizing abnormal keyboard input:
generating Z values of a plurality of dimensions according to the keyboard input behavior, wherein the dimensions refer to the tapping time of each character and/or the interval time of two continuous characters, and each dimension conforms to normal distribution; then, carrying out square summation by using the Z value of each dimension to obtain a chi-square value corresponding to the keyboard input behavior; and finally, judging whether the keyboard input is abnormal or not by comparing the chi-square value with a preset threshold value. The scheme is only suitable for inputting keyboards with fixed contents, such as account numbers, passwords and mail keyboards, but not suitable for inputting ordinary keyboards with any contents and any length.
Disclosure of Invention
The present disclosure provides a keyboard input detection method and apparatus, a storage medium, and an electronic device, which are helpful for detecting abnormal keyboard input without making special requirements on input content, character length, language used, and the like.
In order to achieve the above object, the present disclosure provides a keyboard input detection method, including:
acquiring a keyboard input behavior of a user, wherein the keyboard input behavior comprises an input speed and/or an input content;
obtaining an abnormal score of the keyboard input behavior based on the historical input habit of the user;
and if the abnormal score exceeds a preset score threshold value, judging that the keyboard input behavior belongs to abnormal input.
Optionally, the input speed comprises: the method comprises the following steps of (1) entering the single character at a speed v, a duration T for inputting the character based on the entering speed v, and a frequency n for inputting the character based on the (v, T) occurring in a specified time period T;
the obtaining of the abnormal score of the keyboard input behavior based on the historical input habits of the user comprises:
judging whether the frequency n exceeds a preset frequency n determined based on the historical input habit0；
If the frequency n exceeds the preset frequency n0Then, the probability P of the normal occurrence frequency n in the specified time period T is calculatednT；
Based on the probability PnTObtaining abnormal score S-1-P of the keyboard input behaviornT。
Optionally, the input content includes: a character string entered within a specified time period T, the character string comprising M characters;
the obtaining of the abnormal score of the keyboard input behavior based on the historical input habits of the user comprises:
constructing a language model based on the historical input habits and grammar rules, and predicting a score Q (C) of the character string belonging to normal input by using the language modelT) (ii) a Based on the score Q (C)T) Obtaining abnormal score S-1-Q (C) of the keyboard input behaviorT)；
Or,
extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise syllable segmentation component N of the character string and/or syllable segmentation proportion D of the character string is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT(ii) a Based on the probability PTObtaining abnormal score S-1-P of the keyboard input behaviorT。
Optionally, the input content includes: a character string entered within a specified time period T, the character string comprising M characters;
the obtaining of the abnormal score of the keyboard input behavior based on the historical input habits of the user comprises:
constructing a language model based on the historical input habits and grammar rules, and predicting a score Q (C) of the character string belonging to normal input by using the language modelT)；
Extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise syllable segmentation component N of the character string and/or syllable segmentation proportion D of the character string is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT；
Based on the score Q (C)T) And the probability PTObtaining abnormal score S ═ A (1-Q (C) of the keyboard input behaviorT))+B(1-PT) A, B are empirical parameters.
Optionally, after determining that the keyboard input behavior belongs to abnormal input, the method further includes:
and carrying out abnormal input reminding on the user, and if the user confirms that the keyboard input behavior belongs to normal behavior, determining the keyboard input behavior as the historical input habit.
The present disclosure provides a keyboard input detection apparatus, the apparatus comprising:
the keyboard input behavior acquisition module is used for acquiring the keyboard input behavior of a user, and the keyboard input behavior comprises input speed and/or input content;
the abnormal score obtaining module is used for obtaining the abnormal score of the keyboard input behavior based on the historical input habit of the user;
and the abnormal input judging module is used for judging that the keyboard input behavior belongs to abnormal input when the abnormal score exceeds a preset score threshold value.
Optionally, the input speed comprises: the typing speed v of a single character, the duration T of character input based on the typing speed v, and the frequency n of occurrence of character input based on (v, T) within a specified time period T, then
The abnormal score obtaining module is used for judging whether the frequency n exceeds a preset frequency n determined based on the historical input habit0(ii) a If the frequency n exceeds the preset frequency n0Then, the probability P of the normal occurrence frequency n in the specified time period T is calculatednT(ii) a Based on the probability PnTObtaining abnormal score S-1-P of the keyboard input behaviornT。
Optionally, the input content includes: a character string entered within a specified time period T, the character string comprising M characters;
the abnormal score obtaining module is used for constructing a language model based on the historical input habits and grammar rules, and predicting a score Q (C) of the character string belonging to normal input by using the language modelT) (ii) a Based on the score Q (C)T) Obtaining abnormal score S-1-Q (C) of the keyboard input behaviorT)；
Or,
the abnormal score obtaining module is used for extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise a syllable segmentation component N of the character string and/or a syllable segmentation proportion D of the character string which is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT(ii) a Based on the probability PTObtained byObtaining the abnormal score S-1-P of the keyboard input behaviorT。
Optionally, the input content includes: a character string entered within a specified time period T, the character string including M characters, the abnormality score obtaining module including:
a voice model score prediction module for constructing a language model based on the historical input habits and grammar rules, and predicting the score Q (C) of the character string belonging to normal input by using the language modelT)；
The syllable segmentation characteristic probability calculation module is used for extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise a syllable segmentation component N of the character string and/or a syllable segmentation proportion D of the character string which is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT；
An abnormality score obtaining submodule for obtaining a score Q (C) based on the abnormality scoreT) And the probability PTObtaining abnormal score S ═ A (1-Q (C) of the keyboard input behaviorT))+B(1-PT) A, B are empirical parameters.
Optionally, the apparatus further comprises:
and the normal behavior confirmation module is used for performing abnormal input reminding on the user when the abnormal input judgment module judges that the keyboard input behavior belongs to abnormal input, and determining the keyboard input behavior as the historical input habit if the user confirms that the keyboard input behavior belongs to normal behavior.
The present disclosure provides a storage device having a plurality of instructions stored therein, the instructions being loaded by a processor to perform the steps of the above-described keyboard input detection method.
The present disclosure provides an electronic device, comprising;
the above-mentioned storage device; and
a processor to execute instructions in the storage device.
According to the scheme, when the fact that the user carries out human-computer interaction through the keyboard is monitored, the keyboard input behavior of the user can be obtained, the abnormal score of the keyboard input behavior is calculated based on the historical input habit of the user, and abnormal input detection is achieved. Specifically, if the abnormal score exceeds the preset score threshold, it indicates that the keyboard input behavior does not conform to the historical input habit of the user, and it may be determined that the keyboard input behavior belongs to abnormal input. In addition, as can be seen from the above description, the present disclosure does not require any special restrictions on input contents, character lengths, languages used, and the like, and is applicable to various types of keyboards.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a schematic flow chart of a keyboard input detection method according to an embodiment of the present disclosure;
FIG. 2 is a schematic flow chart illustrating the process of obtaining abnormal scores of keyboard input behaviors in the disclosed embodiment;
FIG. 3 is a schematic diagram of a normal distribution of frequencies in an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of a keyboard input detection device according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of an electronic device for keyboard input detection according to the present disclosure.
Detailed Description
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
Referring to fig. 1, a flow chart of the keyboard input detection method of the present disclosure is shown. May include the steps of:
s101, acquiring a keyboard input behavior of a user, wherein the keyboard input behavior comprises an input speed and/or an input content.
Generally, the input behavior of the user has a certain regularity, that is, the normal keyboard input behavior of the user should be consistent with the past input habits of the user, so when the scheme disclosed by the invention monitors that the user inputs content through the keyboard, the keyboard input behavior of the user can be acquired, and the abnormal input detection is realized based on the historical input habits of the user.
As an example, the keyboard input behavior in the present disclosure may be embodied as an input speed and/or an input content, which is described below in detail at an abnormal score of the keyboard input behavior, and will not be described in detail here.
And S102, obtaining abnormal scores of the keyboard input behaviors based on the historical input habits of the user.
In conjunction with different manifestations of keyboard input behavior, the anomaly score of the disclosed aspects may be embodied in any of the following three cases, each of which is explained below.
1. The keyboard input behavior may be input speed
As an example, the input speed may include: the typing speed v of a single character, the duration T of character input based on the typing speed v, and the frequency n of occurrences of behavior for character input based on (v, T) within a specified time period T.
For example, the typing time of each character can be recorded by an input method, and the typing speed v of a single character is calculated by using the interval between two adjacent typing times, namely the typing speed v belongs to the instantaneous speed. Taking the typing "xiguan" as an example, the typing time t of "x", "i" can be recordedx、tiUsing (t)i-tx) Calculating the typing speed v of the character "xxE.g. (t)i-tx) 0.5s, the typing speed v of the character "xx2 bonds/s; in the same way, (t) can be utilizedg-ti) Calculating the typing speed v of the character "iiBy analogy, the typing speed v corresponding to each character can be obtainedg、vu、va、vn. Understandably, the typing speed v of the last character "nnCan be ignored; or; can be set to a default speed v0(ii) a Or it may be the same as the input speed of the previous character, i.e. vn＝vaThe scheme of the present disclosure may not be specifically limited to this, and may be determined in accordance with practical application requirements.
Again using the key type "xiguan", if vx＝vi＝vg＝v1、vu＝v2Then the velocity v based on the key-in can be calculated1Duration t for character input1. As will be appreciated, if the user types the string "xiguan" in succession, then t1＝(tg-tx) (ii) a If the user has an interruption while typing the character string, for example, after typing "xi", the user performs on-screen candidate to find the corresponding Chinese character "get" and clicks the sending key, etc., the duration t of the interruption can be recordedInterruption of a memoryCorresponding to t1＝(tg-tx-tInterruption of a memory). I.e. t1Based on typing speed v only1The duration of the character input.
As described above, the frequency n of occurrence of the action of character input based on (v, t) can be statistically obtained. For example, if the specified time period T is 30s, (2,2.5,10) it is understood that the typing speed is 2 keys/s and the duration of character input at the typing speed is 2.5s, such a case occurs 10 times in 30 s.
As an example, the duration of the specified time period may be defined according to actual usage requirements, which may not be specifically limited by the present disclosure. Generally, the specified time period does not exceed the acquisition time duration corresponding to the keyboard input behavior.
After obtaining the information related to the input speed, the abnormal score of the keyboard input behavior may be obtained according to the flowchart shown in fig. 2, which may specifically include the following steps:
s201, judging whether the frequency n exceeds a preset frequency n determined based on the historical input habit0。
As an example, the preset frequency n may be determined as follows0：
The probability distribution of all (typing speed and duration) corresponding frequencies under the condition that the total character sample is H can be drawn to represent the historical input habits of the user, and generally, the historical input habits of the user are certain, namely, the historical input habits conform to normal distribution. By way of example, FIG. 3 shows (v)1，t1) And (3) combining a normal distribution diagram of n, wherein the horizontal axis represents the frequency n, and the vertical axis represents the probability of normal occurrence of n.
For example, it can be set that the area of the normal distribution 80% corresponds to the predetermined frequency n0The lowest frequent 10% can be considered as too slow input and the highest frequent 10% as too fast input, both of which can be considered as abnormal input. Specifically, n can be obtained by the following formula0：
The scheme disclosed by the invention can be not limited by the specific expression forms of the area corresponding to the preset frequency, the frequency corresponding to too slow input, the frequency corresponding to too fast input and abnormal input, and can be determined by combining the actual application requirements.
S202, if the frequency n is equal to the preset frequency n0If the two times of the normal occurrence frequency are not consistent, calculating the probability P of the normal occurrence frequency n in the specified time period TnT。
S203, based on the probability PnTObtaining abnormal score S-1-P of the keyboard input behaviornT。
As an example, n and n in the disclosed solution0A discrepancy may be understood as n exceeding the frequency with which the input corresponds too fast, or n being lower than the frequency with which the input corresponds too slow. Taking the example of fig. 3 as an example, if the user is in the process of inputting, n corresponding to (v, t) appears>n0In the case of (1), the user is considered to have performed suspicious abnormal rapid input, and the probability P of the normal occurrence frequency n in the specified time period T can be calculated according to the following formulanTFurther, the score S of the input speed abnormality is obtained as 1-PnTAnd the abnormal score can be used as the abnormal score of the keyboard input behavior.
2. The keyboard input behavior may be inputting content
As an example, inputting content may include: the character string typed in the designated time period T can comprise M characters, and M is more than or equal to 1.
The scheme of the present disclosure provides the following scheme for performing abnormal input detection based on input content:
(1) abnormal input detection based on language model score of character string
As an example, the habit may be entered based on history andthe grammar rule constructs a language model, so that a sentence C obtained by decoding a character string input by a user in a time period T through an input method in a whole sentence way can be obtained after the character string is subjected to the language model, and a score Q (C) of the character string belonging to normal input can be obtainedT) Further, the score S-1-Q (C) indicating the abnormality of the input content is obtainedT) And the abnormal score can be used as the abnormal score of the keyboard input behavior.
Generally, the lower the score, the lower the probability that the user has previously input the character string, and from the perspective of normal grammar rules, the less precedent the character string, the higher the probability that the input content is abnormal.
(2) Abnormal input detection according to syllable segmentation characteristic score of character string
As an example, a character string input by a user in a time period T may be syllable-segmented by a pinyin engine, and a syllable segmentation feature of the character string is extracted, for example, the syllable segmentation feature may include a syllable segmentation component N of the character string, and/or a syllable segmentation ratio D of the character string is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on historical input habits or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, the probability P of normally obtaining the syllable segmentation characteristic can be calculatedTFurther, the score S of the input content abnormality is obtained as 1-PTAnd the abnormal score can be used as the abnormal score of the keyboard input behavior.
Generally, when a user inputs normally, the input content has a certain regularity. For example, a user is accustomed to short spellings, as a single input may only involve one or two syllables; or, the user is used to conciliate, for example, only the initial consonant or the initial letter of the initial consonant is input, and the syllable splitting ratio is close to 1: 1; or, the user input has personalized characteristics, such as partially abbreviating the phrases, taking "working" as an example, and the input mode of the user habit is "shangb".
For example, if "dhusahfajbahda" is input at the keyboard, the following syllables may be cut: d 'hu' sa 'h' fa 'j' ba 'h' da, i.e. 14 characters are cut into 9 syllables, which usually do not conform to the input habit of the user and can be recognized as abnormal input.
Corresponding to the syllable segmentation proportion, the preset segmentation characteristics can be determined according to the following method:
the method can carry out data accumulation learning on the syllable segmentation proportion input by the user at a single time, and establish corresponding probability distribution functions F (D) aiming at different syllable segmentation proportions. As an example, the preset segmentation feature D can be determined by utilizing a syllable segmentation ratio corresponding to 80% probability0Specifically, D can be obtained by the following formula0：
If the user is in the process of inputting, D appears>D0In the case of (3), the user may be considered to have performed suspicious abnormal input, and the probability P of obtaining D by normal segmentation within the specified time period T may be determined based on the probability distribution function F (D)TFurther, the score S of the input content abnormality is obtained as 1-PTAnd the abnormal score can be used as the abnormal score of the keyboard input behavior.
Taking the syllable-dividing component as an example, the syllable-dividing component input by the user at a single time can be counted, and the corresponding probability distribution function F (M) is established for different syllable-dividing components. Similar to the syllable segmentation ratio, the preset segmentation characteristic M can be determined0And is in M>M0Then, the score S of the input content abnormality is calculated to be 1-PTAs an abnormal score for the keyboard input behavior.
Understandably, compute D0、M0The probability of the time selection can be determined by combining with the actual application requirements, and the scheme disclosed by the invention can not be specifically limited.
(3) Abnormal input detection based on language model score and syllable segmentation feature score of character string
As described above, post-Q (C) is obtainedT)、PTThe abnormal score S ═ a (1-Q (C) of the keyboard input behavior can be obtained based on both of themT))+B(1-PT) A, B are empirical parameters. The specific process can be described with reference to the above description, and is not further described herein.
3. The keyboard input behavior can be input speed and input content
As described above, the score (1-P) of the input speed anomaly is obtainednT) And scores of input content anomalies, e.g. [ A (1-Q (C) ]T))+B(1-PT)]The abnormal score S ═ a (1-P) of the keyboard input behavior can be obtained based on the twonT)+b[A(1-Q(CT))+B(1-PT)]And a and b are empirical parameters. The specific process can be described with reference to the above description, and is not further described herein.
S103, if the abnormal score exceeds a preset score threshold value, judging that the keyboard input behavior belongs to abnormal input.
In conjunction with the above three cases of anomaly scores, the preset score threshold may be embodied as:
1. abnormal score S-1-P of keyboard input behaviornT
The preset score threshold S can be determined by combining the input speed of the past keyboard input behaviors of the user1If S is>S1Then, the keyboard input behavior can be determined to be abnormal input.
2. Abnormal score of keyboard input behavior S ═ a (1-Q (C)T))+B(1-PT)
The preset score threshold S can be determined by combining the input content of the past keyboard input behaviors of the user2If S is>S2Then, the keyboard input behavior can be determined to be abnormal input.
3. Abnormal score of keyboard input behavior S ═ a (1-P)nT)+b[A(1-Q(CT))+B(1-PT)]
Input speed capable of combining with past keyboard input behaviors of userAnd inputting the content, determining a preset score threshold S3If S is>S3Then, the keyboard input behavior can be determined to be abnormal input.
In summary, the present disclosure may obtain the keyboard input behavior of the user, and calculate the abnormal score of the keyboard input behavior, generally, the lower the abnormal score is, the more the keyboard input behavior conforms to the historical input habit of the user, and the higher the probability that the keyboard input behavior belongs to normal input is; otherwise, the probability of belonging to an abnormal input is higher. In addition, as can be seen from the above description, the present disclosure does not require special restrictions on input content, character length, language used, and the like, and is applicable to various types of keyboards, for example, a special keyboard dedicated to account and password input, or a common keyboard used in daily life; alternatively, it may be a physical keyboard or a virtual keyboard.
It is understood that the relevant threshold referred to in the present disclosure may be a personalized threshold determined based on the current user's historical keyboard input behavior; alternatively, the threshold may be a general threshold determined based on historical keyboard input behaviors of most users, which may not be specifically limited by the present disclosure.
As an example, the present disclosure further provides an optimization scheme of adaptive learning, specifically, after determining that the keyboard input behavior of the user belongs to abnormal input, an abnormal input prompt may be performed to the user, and if the user confirms that the keyboard input behavior belongs to normal behavior, the keyboard input behavior may be determined as a historical input habit of the user, and is used to optimize a threshold related to input speed and/or input content, and the like. By the iterative optimization scheme, the accuracy of abnormal input detection of the scheme can be improved, and the false detection rate is reduced. In addition, if the user confirms that the keyboard input behavior belongs to the abnormal behavior, as an example, the character string input by the keyboard input behavior may be cleared at one time.
As an example, after determining that the keyboard input behavior of the user belongs to the abnormal input, the subsequent input characters of the user can be continuously received; alternatively, subsequent typing by the user may be masked and processed accordingly as selected by the user. The present disclosure may not be particularly limited with respect to a processing procedure after the keyboard input behavior is determined as the abnormal input.
Referring to fig. 4, a schematic diagram of the keyboard input detection device of the present disclosure is shown. The apparatus may include:
a keyboard input behavior acquisition module 301, configured to acquire a keyboard input behavior of a user, where the keyboard input behavior includes an input speed and/or an input content;
an abnormal score obtaining module 302, configured to obtain an abnormal score of the keyboard input behavior based on a historical input habit of a user;
an abnormal input determination module 303, configured to determine that the keyboard input behavior belongs to abnormal input when the abnormal score exceeds a preset score threshold.
Optionally, the input speed comprises: the typing speed v of a single character, the duration T of character input based on the typing speed v, and the frequency n of occurrence of character input based on (v, T) within a specified time period T, then
The abnormal score obtaining module is used for judging whether the frequency n exceeds a preset frequency n determined based on the historical input habit0(ii) a If the frequency n exceeds the preset frequency n0Then, the probability P of the normal occurrence frequency n in the specified time period T is calculatednT(ii) a Based on the probability PnTObtaining abnormal score S-1-P of the keyboard input behaviornT。
Optionally, the input content includes: a character string entered within a specified time period T, the character string comprising M characters;
the abnormal score obtaining module is used for constructing a language model based on the historical input habits and the grammar rules and utilizing the language modelType predicts the score Q (C) of the character string belonging to normal inputT) (ii) a Based on the score Q (C)T) Obtaining abnormal score S-1-Q (C) of the keyboard input behaviorT)；
Or,
the abnormal score obtaining module is used for extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise a syllable segmentation component N of the character string and/or a syllable segmentation proportion D of the character string which is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT(ii) a Based on the probability PTObtaining abnormal score S-1-P of the keyboard input behaviorT。
Optionally, the input content includes: a character string entered within a specified time period T, the character string including M characters, the abnormality score obtaining module including:
a voice model score prediction module for constructing a language model based on the historical input habits and grammar rules, and predicting the score Q (C) of the character string belonging to normal input by using the language modelT)；
The syllable segmentation characteristic probability calculation module is used for extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise a syllable segmentation component N of the character string and/or a syllable segmentation proportion D of the character string which is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT；
An abnormality score obtaining submodule for obtaining a score Q (C) based on the abnormality scoreT) And the probability PTObtaining the input line of the keyboardAbnormal score of (A) is (1-Q (C)T))+B(1-PT) A, B are empirical parameters.
Optionally, the apparatus further comprises:
and the normal behavior confirmation module is used for performing abnormal input reminding on the user when the abnormal input judgment module judges that the keyboard input behavior belongs to abnormal input, and determining the keyboard input behavior as the historical input habit if the user confirms that the keyboard input behavior belongs to normal behavior.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Referring to fig. 5, a schematic structural diagram of an electronic device 400 for keyboard input detection according to the present disclosure is shown. Referring to fig. 5, electronic device 400 includes a processing component 401 that further includes one or more processors, and storage resources, represented by storage medium 402, for storing instructions, such as application programs, that are executable by processing component 401. The application stored in the storage medium 402 may include one or more modules that each correspond to a set of instructions. Further, the processing component 401 is configured to execute instructions to perform the keyboard input detection method described above.
Electronic device 400 may also include a power component 403 configured to perform power management of electronic device 400; a wired or wireless network interface 404 configured to connect the electronic device 400 to a network; and an input/output (I/O) interface 405. The electronic device 400 may operate based on an operating system stored on the storage medium 402, such as WindowsServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, or the like.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that, in the foregoing embodiments, various features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various combinations that are possible in the present disclosure are not described again.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.
Claims (12)
1. A keyboard input detection method, the method comprising:
acquiring a keyboard input behavior of a user, wherein the keyboard input behavior comprises an input speed and/or an input content;
obtaining an abnormal score of the keyboard input behavior based on the historical input habit of the user;
and if the abnormal score exceeds a preset score threshold value, judging that the keyboard input behavior belongs to abnormal input.
2. The method of claim 1, wherein the input speed comprises: the method comprises the following steps of (1) entering the single character at a speed v, a duration T for inputting the character based on the entering speed v, and a frequency n for inputting the character based on the (v, T) occurring in a specified time period T;
the obtaining of the abnormal score of the keyboard input behavior based on the historical input habits of the user comprises:
judging whether the frequency n exceeds a preset frequency n determined based on the historical input habit0；
If the frequency n exceeds the preset frequency n0Then, the probability P of the normal occurrence frequency n in the specified time period T is calculatednT；
Based on the probability PnTObtaining abnormal score S-1-P of the keyboard input behaviornT。
3. The method of claim 1, wherein the inputting the content comprises: a character string entered within a specified time period T, the character string comprising M characters;
the obtaining of the abnormal score of the keyboard input behavior based on the historical input habits of the user comprises:
constructing a language model based on the historical input habits and grammar rules, and predicting a score Q (C) of the character string belonging to normal input by using the language modelT) (ii) a Based on the score Q (C)T) Obtaining abnormal score S-1-Q (C) of the keyboard input behaviorT)；
Or,
extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise syllable segmentation component N of the character string and/or syllable segmentation proportion D of the character string is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT(ii) a Based on the probability PTObtaining abnormal score S-1-P of the keyboard input behaviorT。
4. The method of claim 1, wherein the inputting the content comprises: a character string entered within a specified time period T, the character string comprising M characters;
the obtaining of the abnormal score of the keyboard input behavior based on the historical input habits of the user comprises:
constructing a language model based on the historical input habits and grammar rules, and predicting a score Q (C) of the character string belonging to normal input by using the language modelT)；
Extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise syllable segmentation component N of the character string and/or syllable segmentation proportion D of the character string is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT；
Based on the score Q (C)T) And the probability PTObtaining abnormal score S ═ A (1-Q (C) of the keyboard input behaviorT))+B(1-PT) A, B are empirical parameters.
5. The method of any of claims 1-4, wherein after determining that the keyboard input behavior pertains to abnormal input, the method further comprises:
and carrying out abnormal input reminding on the user, and if the user confirms that the keyboard input behavior belongs to normal behavior, determining the keyboard input behavior as the historical input habit.
6. A keyboard input detection apparatus, comprising:
the keyboard input behavior acquisition module is used for acquiring the keyboard input behavior of a user, and the keyboard input behavior comprises input speed and/or input content;
the abnormal score obtaining module is used for obtaining the abnormal score of the keyboard input behavior based on the historical input habit of the user;
and the abnormal input judging module is used for judging that the keyboard input behavior belongs to abnormal input when the abnormal score exceeds a preset score threshold value.
7. The apparatus of claim 6, wherein the input speed comprises: the typing speed v of a single character, the duration T of character input based on the typing speed v, and the frequency n of occurrence of character input based on (v, T) within a specified time period T, then
The abnormal score obtaining module is used for judging whether the frequency n exceeds a preset frequency n determined based on the historical input habit0(ii) a If the frequency n exceeds the preset frequency n0Then, the probability P of the normal occurrence frequency n in the specified time period T is calculatednT(ii) a Based on the probability PnTObtaining abnormal score S-1-P of the keyboard input behaviornT。
8. The apparatus of claim 6, wherein the input content comprises: a character string entered within a specified time period T, the character string comprising M characters;
the abnormal score obtaining module is used for constructing a language model based on the historical input habits and grammar rules, and predicting a score Q (C) of the character string belonging to normal input by using the language modelT) (ii) a Based on the score Q (C)T) Obtaining abnormal score S-1-Q (C) of the keyboard input behaviorT)；
Or,
the abnormal score obtaining module is used for extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise syllable segmentation component N and ^ whether or not the syllable segmentation component N is greater than or equal toOr the syllable segmentation proportion D of the character string is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT(ii) a Based on the probability PTObtaining abnormal score S-1-P of the keyboard input behaviorT。
9. The apparatus of claim 6, wherein the input content comprises: a character string entered within a specified time period T, the character string including M characters, the abnormality score obtaining module including:
a voice model score prediction module for constructing a language model based on the historical input habits and grammar rules, and predicting the score Q (C) of the character string belonging to normal input by using the language modelT)；
The syllable segmentation characteristic probability calculation module is used for extracting syllable segmentation characteristics of the character string, wherein the syllable segmentation characteristics comprise a syllable segmentation component N of the character string and/or a syllable segmentation proportion D of the character string which is N/M; judging whether the syllable segmentation characteristic exceeds a preset segmentation characteristic determined based on the historical input habit or not; if the syllable segmentation characteristic exceeds the preset segmentation characteristic, calculating the probability P of normally obtaining the syllable segmentation characteristic in the specified time period TT；
An abnormality score obtaining submodule for obtaining a score Q (C) based on the abnormality scoreT) And the probability PTObtaining abnormal score S ═ A (1-Q (C) of the keyboard input behaviorT))+B(1-PT) A, B are empirical parameters.
10. The apparatus of any one of claims 6 to 9, further comprising:
and the normal behavior confirmation module is used for performing abnormal input reminding on the user when the abnormal input judgment module judges that the keyboard input behavior belongs to abnormal input, and determining the keyboard input behavior as the historical input habit if the user confirms that the keyboard input behavior belongs to normal behavior.
11. A storage device having stored therein a plurality of instructions, wherein said instructions are loaded by a processor for performing the steps of the method of any of claims 1 to 5.
12. An electronic device, characterized in that the electronic device comprises;
the storage device of claim 11; and
a processor to execute instructions in the storage device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711418534.6A CN108182000A (en) | 2017-12-25 | 2017-12-25 | Keyboard input detection method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711418534.6A CN108182000A (en) | 2017-12-25 | 2017-12-25 | Keyboard input detection method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108182000A true CN108182000A (en) | 2018-06-19 |
Family
ID=62547253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711418534.6A Pending CN108182000A (en) | 2017-12-25 | 2017-12-25 | Keyboard input detection method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108182000A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033205A (en) * | 2018-06-29 | 2018-12-18 | 上海精密计量测试研究所 | Model is checked based on the Aerospace test data that data input columns are analysis |
| CN109873813A (en) * | 2019-01-28 | 2019-06-11 | 平安科技（深圳）有限公司 | Text input abnormal monitoring method, device, computer equipment and storage medium |
| CN110263337A (en) * | 2019-06-17 | 2019-09-20 | 北京百度网讯科技有限公司 | Input behavior recognition methods, device, equipment and medium |
| CN110536017A (en) * | 2019-09-03 | 2019-12-03 | 中国联合网络通信集团有限公司 | Message sending control method, message transmission control device |
| CN110865716A (en) * | 2019-09-29 | 2020-03-06 | 上海萌家网络科技有限公司 | Input method and system thereof |
| CN113971271A (en) * | 2020-07-22 | 2022-01-25 | 华为技术有限公司 | Fingerprint unlocking method and device, terminal and storage medium |
| WO2023108833A1 (en) * | 2021-12-15 | 2023-06-22 | 三六零科技集团有限公司 | Terminal anomalous behavior detection method and apparatus, device, and storage medium |
| CN117251095A (en) * | 2023-09-12 | 2023-12-19 | 深圳市驿格科技有限公司 | Data input method and system for PDA |
| CN118094532A (en) * | 2024-04-29 | 2024-05-28 | 曙光信息产业(北京)有限公司 | Intelligent protection method for processing storage hardware |
| CN119621398A (en) * | 2024-11-26 | 2025-03-14 | 广州迈派电子有限公司 | Temporary replacement input method and system for bad keys of keyboard and mouse |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102193736A (en) * | 2011-04-21 | 2011-09-21 | 安徽科大讯飞信息科技股份有限公司 | Input method and system supporting multimode automatic switching |
| CN102622095A (en) * | 2011-01-03 | 2012-08-01 | 仁宝电脑工业股份有限公司 | Method for preventing touch keyboard from being accidentally touched and touch electronic device |
| CN103105942A (en) * | 2013-03-01 | 2013-05-15 | 王湘民 | Method for simplifying Chinese-foreign language information input and judging whether input is right or not |
| EP2646904A1 (en) * | 2010-11-29 | 2013-10-09 | Biocatch Ltd. | Method and device for confirming computer end-user identity |
| CN105243301A (en) * | 2014-07-09 | 2016-01-13 | 阿里巴巴集团控股有限公司 | Keyboard input abnormality detection method and apparatus as well as security prompt method and apparatus |
| CN107092424A (en) * | 2016-02-18 | 2017-08-25 | 北京搜狗科技发展有限公司 | A kind of display methods of error correction, device and the device of the display for error correction |
-
2017
- 2017-12-25 CN CN201711418534.6A patent/CN108182000A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2646904A1 (en) * | 2010-11-29 | 2013-10-09 | Biocatch Ltd. | Method and device for confirming computer end-user identity |
| CN102622095A (en) * | 2011-01-03 | 2012-08-01 | 仁宝电脑工业股份有限公司 | Method for preventing touch keyboard from being accidentally touched and touch electronic device |
| CN102193736A (en) * | 2011-04-21 | 2011-09-21 | 安徽科大讯飞信息科技股份有限公司 | Input method and system supporting multimode automatic switching |
| CN103105942A (en) * | 2013-03-01 | 2013-05-15 | 王湘民 | Method for simplifying Chinese-foreign language information input and judging whether input is right or not |
| CN105243301A (en) * | 2014-07-09 | 2016-01-13 | 阿里巴巴集团控股有限公司 | Keyboard input abnormality detection method and apparatus as well as security prompt method and apparatus |
| CN107092424A (en) * | 2016-02-18 | 2017-08-25 | 北京搜狗科技发展有限公司 | A kind of display methods of error correction, device and the device of the display for error correction |
Non-Patent Citations (2)
| Title |
|---|
| 乔立岩，等: "基于支持向量机的键盘密码输入异常检测方法研究", 《电子测量与仪器学报》 * |
| 饶志宏 等 著: "《工业SCADA系统信息安全技术》", 31 May 2014, 国防工业出版社 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033205B (en) * | 2018-06-29 | 2021-12-31 | 上海精密计量测试研究所 | Aerospace test data checking device based on data input behavior analysis |
| CN109033205A (en) * | 2018-06-29 | 2018-12-18 | 上海精密计量测试研究所 | Model is checked based on the Aerospace test data that data input columns are analysis |
| CN109873813A (en) * | 2019-01-28 | 2019-06-11 | 平安科技（深圳）有限公司 | Text input abnormal monitoring method, device, computer equipment and storage medium |
| WO2020155773A1 (en) * | 2019-01-28 | 2020-08-06 | 平安科技（深圳）有限公司 | Method of monitoring for suspicious text input, device, computer apparatus, and storage medium |
| CN110263337A (en) * | 2019-06-17 | 2019-09-20 | 北京百度网讯科技有限公司 | Input behavior recognition methods, device, equipment and medium |
| CN110263337B (en) * | 2019-06-17 | 2023-06-20 | 北京百度网讯科技有限公司 | Input behavior recognition method, device, equipment and medium |
| CN110536017A (en) * | 2019-09-03 | 2019-12-03 | 中国联合网络通信集团有限公司 | Message sending control method, message transmission control device |
| CN110865716A (en) * | 2019-09-29 | 2020-03-06 | 上海萌家网络科技有限公司 | Input method and system thereof |
| CN113971271A (en) * | 2020-07-22 | 2022-01-25 | 华为技术有限公司 | Fingerprint unlocking method and device, terminal and storage medium |
| WO2023108833A1 (en) * | 2021-12-15 | 2023-06-22 | 三六零科技集团有限公司 | Terminal anomalous behavior detection method and apparatus, device, and storage medium |
| CN117251095A (en) * | 2023-09-12 | 2023-12-19 | 深圳市驿格科技有限公司 | Data input method and system for PDA |
| CN117251095B (en) * | 2023-09-12 | 2024-05-17 | 深圳市驿格科技有限公司 | Data input method and system for PDA |
| CN118094532A (en) * | 2024-04-29 | 2024-05-28 | 曙光信息产业(北京)有限公司 | Intelligent protection method for processing storage hardware |
| CN119621398A (en) * | 2024-11-26 | 2025-03-14 | 广州迈派电子有限公司 | Temporary replacement input method and system for bad keys of keyboard and mouse |
| CN119621398B (en) * | 2024-11-26 | 2025-09-26 | 广州迈派电子有限公司 | Bad key temporary substitution input method and system for key mouse |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108182000A (en) | Keyboard input detection method and device, storage medium and electronic equipment | |
| JP6594534B2 (en) | Text information processing method and device | |
| CN105814519B (en) | System and method for inputting image or label to electronic equipment | |
| US8542195B2 (en) | Method for optimization of soft keyboards for multiple languages | |
| CN107544684B (en) | Candidate word display method and device | |
| CN107608532A (en) | A kind of association-feeding method, device and electronic equipment | |
| WO2014159473A2 (en) | Automatic supplementation of word correction dictionaries | |
| CN107155121B (en) | Voice control text display method and device | |
| CN109002183B (en) | Information input method and device | |
| US8830182B1 (en) | Keystroke resolution | |
| WO2022083750A1 (en) | Text display method and apparatus and electronic device | |
| CN107291260B (en) | Information input method and device for inputting information | |
| CN108268438B (en) | Page content extraction method and device and client | |
| US10614796B2 (en) | Method of and system for processing a user-generated input command | |
| EP3485361B1 (en) | Pressure-based gesture typing for a graphical keyboard | |
| CN104076945B (en) | Apparatus and method for showing input keyboard in the terminal | |
| CN109543014B (en) | Man-machine conversation method, device, terminal and server | |
| CN109062891B (en) | Media processing method, device, terminal and medium | |
| Kryvonos et al. | Predictive text typing system for the Ukrainian language | |
| US20160092104A1 (en) | Methods, systems and devices for interacting with a computing device | |
| CN107797676A (en) | A kind of input method of the single character and device | |
| CN103389800B (en) | A kind of method and apparatus generating entry | |
| CN112558784A (en) | Method and device for inputting characters and electronic equipment | |
| CN106156022B (en) | Information processing method and electronic equipment | |
| CN111103986B (en) | User word stock management method and device, and user word stock input method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180730 Address after: 310000 A-B102-930 room 198, road, Xiaoshan economic and Technological Development Zone, Xiaoshan District, Hangzhou. Applicant after: Zhejiang flying Intelligent Technology Co., Ltd. Address before: 230000 666 Wangjiang West Road, Hefei hi tech Development Zone, Anhui Applicant before: Iflytek Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180619 |
|
| RJ01 | Rejection of invention patent application after publication |