CN110232132B - Time series data processing method and device - Google Patents
Time series data processing method and device Download PDFInfo
- Publication number
- CN110232132B CN110232132B CN201910526163.6A CN201910526163A CN110232132B CN 110232132 B CN110232132 B CN 110232132B CN 201910526163 A CN201910526163 A CN 201910526163A CN 110232132 B CN110232132 B CN 110232132B
- Authority
- CN
- China
- Prior art keywords
- time point
- state
- observation window
- energy
- fluctuation energy
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
Abstract
The invention discloses a time sequence data processing method and a time sequence data processing device, wherein the method comprises the following steps: acquiring time series data, wherein each data in the time series is a value of a time point; sequentially calculating fluctuation energy e (n) ═ V (n) — V (n-1) at each time point, where e (n) is the fluctuation energy at the current time point, V (n) is the value at the current time point, and V (n-1) is the value at the previous time point; circularly traversing each time point in the time sequence, and determining the state of each time point according to the fluctuation energy of each time point, wherein the state comprises the following steps: transient state and steady state; taking the stable state of the time point as the state mode of the time point; the status patterns include: a steady mode, an up mode, a down mode. By utilizing the invention, the segmentation of the time series data and the identification of the state mode can be simply and accurately realized.
Description
Technical Field
The invention relates to the field of data processing, in particular to a time series data processing method and device.
Background
Time series data refers to data collected in chronological order, such data reflecting the state or degree of change of a certain thing, phenomenon, etc. with time, which can be visually represented in a plane as a time series waveform such as an Electrocardiogram (ECG), an electroencephalogram (EEG), a current-voltage signal in manufacturing, a K-line of stock trading, a time-domain waveform of a voice signal, etc. In such data analysis applications, it is generally necessary to segment the waveform and classify the segmented waveform to provide corresponding information for various applications.
In the prior art, there are various ways to segment time series data, but most of the prior art have various limitations or drawbacks, such as: the method has the advantages of high computational complexity, inaccurate segmentation in a complex scene, incapability of being suitable for real-time computation, large influence by noise data, strong hypothesis on the distribution or source of the data, requirement of training data and labels for supervised learning, suitability for certain signals with specific graph patterns, and the like.
Disclosure of Invention
The embodiment of the invention provides a time series data processing method and device, and aims to solve one or more problems in the prior art.
Therefore, the invention provides the following technical scheme:
a method of time series data processing, the method comprising:
sampling a time sequence diagram to be processed to obtain time sequence data, wherein each data in the time sequence is a value of a time point, and the time sequence diagram is any one or more of the following: electrocardiogram, electroencephalogram, current-voltage signal in production and manufacture, K line of stock exchange, and time domain waveform of voice signal;
sequentially calculating fluctuation energy e (n) ═ V (n) — V (n-1) at each time point, where e (n) is the fluctuation energy at the current time point, V (n) is the value at the current time point, and V (n-1) is the value at the previous time point;
circularly traversing each time point in the time sequence, and determining the state of each time point according to the fluctuation energy of each time point, wherein the state comprises the following steps: transient state and steady state;
taking the stable state of the time point as the state mode of the time point; the status patterns include: a steady mode, an ascending mode, a descending mode;
and outputting the change point of the time series according to the state mode of each time point in the time series data.
Optionally, the cyclically traversing each time point in the time series, and determining the state of the time point according to the fluctuation energy of each time point includes:
circularly traversing each time point in the time sequence, and determining the temporary state of the time point according to the fluctuation energy of the time point;
and determining the stable state of the time point according to the absolute value and the temporary state of the fluctuation energy of the time point.
Optionally, the determining the temporal state of the time point according to the fluctuation energy of the time point includes:
if the fluctuation energy E (n) >0 at the current time point, determining that the temporary state St (n) at the current time point is in the ascending mode;
determining that the temporary state St (n) of the current time point is a descending mode if the fluctuation energy E (n) of the current time point is less than 0;
if the fluctuation energy e (n) at the current time point is 0, it is determined that the temporal state st (n) at the current time point is a stationary mode.
Optionally, the determining the steady state of the time point according to the absolute value of the fluctuation energy of the time point and the transient state includes:
if the absolute value abs (E (n)) of the fluctuation energy at the current time point is greater than or equal to the set fluctuation energy threshold, taking the temporary state St (n) at the current time point as the stable state at the current time point;
if the absolute value abs (e (n)) of the fluctuation energy at the present time point is smaller than the fluctuation energy threshold, and the transient state St (n) at the present time point is the same as the transient state St (n-1) at the previous time point, the last steady state is taken as the steady state at the present time point;
if the absolute value abs (E (n)) of the fluctuation energy at the current time point is smaller than the fluctuation energy threshold and the temporal state St (n) at the current time point is different from the temporal state St (n-1) at the previous time point, performing unsteady state tracking using a dynamic observation window, and determining the steady state at each time point within the observation window.
Optionally, the performing non-steady state tracking by using the dynamic observation window, and determining a steady state at the current time point includes:
opening a dynamic observation window;
recording the length of the observation window and calculating the accumulated energy sum (E) and the absolute value abs (sum (E)) of the accumulated energy at all time points within the observation window;
judging whether the observation window meets a termination condition;
if yes, terminating the observation window, and determining the stable state of each time point in the observation window according to the termination condition.
Optionally, the termination condition comprises any one of:
a specific time point appears in the observation window, wherein the specific time point refers to a time point of which the absolute value abs (E (i)) of the fluctuation energy is greater than or equal to the fluctuation energy threshold;
the observation window reaches a set length threshold value w;
the absolute value abs (sum (E)) of the accumulated energy of all time points in the observation window is greater than or equal to the fluctuation energy threshold;
the determining the stable state of each time point in the observation window according to the termination condition includes:
in the event that the specific point in time within the observation window occurs to terminate the observation window: moving the termination point of the observation window to the previous time point of the specific time point to obtain an updated observation window; if the length of the updated observation window is greater than half of the set length threshold value w, setting the stable state of each time point in the updated observation window as the last stable state, otherwise, setting the stable state of each time point in the updated observation window as a stable mode;
in case the observation window reaches a set length threshold w terminating the observation window: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0; setting a steady state at each time point within the observation window to a steady mode if an absolute value abs (sum (e)) of the accumulated energy is equal to or less than the fluctuation energy threshold;
in a case where the absolute value abs (sum (e)) of the accumulated energy at all time points within the observation window is equal to or greater than the fluctuation energy threshold value, the observation window is terminated: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0.
Optionally, the sampling the timing graph includes:
the timing diagram is sampled periodically or the timing diagram is sampled aperiodically.
Optionally, the method further comprises:
and displaying the state mode of each time point in the time sequence data.
A time-series data processing apparatus, the apparatus comprising:
the data acquisition module is configured to sample a time sequence diagram to be processed to obtain time sequence data, where each data in the time sequence is a value at a time point, and the time sequence diagram is any one or more of the following: electrocardiogram, electroencephalogram, current-voltage signal in production and manufacture, K line of stock exchange, and time domain waveform of voice signal;
the fluctuation energy calculation module is used for sequentially calculating fluctuation energy E (n) ═ V (n) — V (n-1) of each time point, wherein E (n) is the fluctuation energy of the current time point, V (n) is the value of the current time point, and V (n-1) is the value of the previous time point;
a traversing module, configured to cycle through each time point in the time sequence, and determine a state of each time point according to fluctuation energy of each time point, where the state includes: transient state and steady state;
the output module is used for taking the stable state of the time point as the state mode of the time point and outputting the change point of the time sequence according to the state mode of each time point in the time sequence data; the status patterns include: a steady mode, an up mode, a down mode.
Optionally, the traversing module includes:
the temporary state determining module is used for circularly traversing each time point in the time sequence and determining the temporary state of the time point according to the fluctuation energy of the time point;
and the steady state determining module is used for determining the steady state of the time point according to the absolute value of the fluctuation energy of the time point and the temporary state.
Optionally, the temporary state determining module is specifically configured to determine that the temporary state st (n) at the current time point is an ascending mode when the fluctuation energy e (n) >0 at the current time point; when the fluctuation energy E (n) at the current time point is less than 0, determining that the temporary state St (n) at the current time point is a descending mode; when the fluctuation energy e (n) at the current time point is 0, the temporal state st (n) at the current time point is determined to be the stationary mode.
Optionally, the steady state determination module includes:
a determination unit for determining whether or not the absolute value abs (e (n)) of the fluctuation energy at the current time point is equal to or greater than a set fluctuation energy threshold, and whether or not the provisional state St (n) at the current time point is the same as the provisional state St (n-1) at the previous time point;
a first steady-state determination unit configured to take a temporary state st (n) at a current time point as a steady state at the current time point when an absolute value abs (e (n)) of fluctuation energy at the current time point is equal to or greater than a set fluctuation energy threshold;
a second steady-state determination unit configured to take the last steady state as the steady state at the current time point when the absolute value abs (e (n)) of the fluctuation energy at the current time point is smaller than the fluctuation energy threshold and the transient state St (n) at the current time point is the same as the transient state St (n-1) at the previous time point;
and a third steady-state determination unit configured to perform unsteady-state tracking using a dynamic observation window and determine a steady state at each time point within the observation window when an absolute value abs (e (n)) of fluctuation energy at a current time point is smaller than the fluctuation energy threshold and a transient state St (n) at the current time point is different from a transient state St (n-1) at a previous time point.
Optionally, the third steady-state determining unit is specifically configured to open a dynamic observation window, record the length of the observation window, calculate the accumulated energy sum (e) and the absolute value abs (sum (e)) of the accumulated energy at all time points in the observation window, terminate the observation window after the observation window meets a termination condition, and determine a steady state of each time point in the observation window according to the termination condition.
Optionally, the termination condition comprises any one of:
a specific time point appears in the observation window, wherein the specific time point refers to a time point of which the absolute value abs (E (i)) of the fluctuation energy is greater than or equal to the fluctuation energy threshold;
the observation window reaches a set length threshold value w;
the absolute value abs (sum (E)) of the accumulated energy of all time points in the observation window is greater than or equal to the fluctuation energy threshold;
the third steady-state determination unit determines the steady state at each time point within the observation window in the following manner:
in the event that the specific point in time within the observation window occurs to terminate the observation window: moving the termination point of the observation window to the previous time point of the specific time point to obtain an updated observation window; if the length of the updated observation window is greater than half of the set length threshold value w, setting the stable state of each time point in the updated observation window as the last stable state, otherwise, setting the stable state of each time point in the updated observation window as a stable mode;
in case the observation window reaches a set length threshold w terminating the observation window: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0; setting a steady state at each time point within the observation window to a steady mode if an absolute value abs (sum (e)) of the accumulated energy is equal to or less than the fluctuation energy threshold;
in a case where the absolute value abs (sum (e)) of the accumulated energy at all time points within the observation window is equal to or greater than the fluctuation energy threshold value, the observation window is terminated: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0.
Optionally, the sampling the time-series graph by the data acquisition module includes: the timing diagram is sampled periodically or the timing diagram is sampled aperiodically.
Optionally, the apparatus further comprises:
and the display module is used for displaying the state mode of each time point in the time sequence data.
An electronic device, comprising: one or more processors, memory;
the memory is configured to store computer-executable instructions and the processor is configured to execute the computer-executable instructions to implement the method described above.
A readable storage medium having stored thereon instructions which are executed to implement the foregoing method.
According to the time series data processing method and device provided by the embodiment of the invention, the fluctuation energy of each time point is calculated and calculated aiming at the time series data, and the state mode of each time point is determined by circularly traversing each time point in the time series by using the fluctuation energy of each time point.
Further, the division of a stable state and a non-stable state is proposed, and in the traversal process, the temporary state of each time point is determined according to the fluctuation energy of each time point; and then determining the stable state of the time point according to the absolute value and the temporary state of the fluctuation energy of the time point, so as to obtain the state mode of each time point.
Further, when the fluctuation energy of the current time point is small, if the transient state of the current time point is different from the transient state of the previous time point, the dynamic observation window is used for performing unsteady state tracking to determine the steady state of each time point in the observation window, that is, in the case that the state pattern of the current time point cannot be determined only according to the fluctuation energy of the current time point and the transient state of the previous time point, the state pattern of the current time point is finally determined by considering the transient state conditions of one or more time points after the current time point, so that the determination result of the state pattern of each time point can be more accurate, the continuation of the transient state can be accurately and effectively divided, and the improper division of the state pattern with long time persistence due to the transient change of the energy direction can be reduced. In addition, the influence of energy accumulation change with a certain length or time on the current time point can be accurately judged through unsteady state tracking, and the accuracy of state mode judgment is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings can be obtained by those skilled in the art according to the drawings.
FIG. 1 is a flow chart of a time series data processing method according to an embodiment of the present invention;
FIG. 2 is a flow chart of an embodiment of the present invention for performing unsteady state tracking using a dynamic observation window;
FIG. 3 is a block diagram showing a configuration of a time-series data processing apparatus according to an embodiment of the present invention;
fig. 4 is another block diagram of the time-series data processing apparatus according to the embodiment of the present invention.
Detailed Description
In order to make the technical field of the invention better understand the scheme of the embodiment of the invention, the embodiment of the invention is further described in detail with reference to the drawings and the implementation mode.
The embodiment of the invention provides a time sequence data processing method and a time sequence data processing device, wherein the method comprises the steps of calculating the fluctuation energy of each time point in the time sequence data, circularly traversing each time point in the time sequence by using the fluctuation energy of each time point, and determining the state of each time point according to the fluctuation energy of each time point, wherein the state comprises the following steps: transient state and steady state; and taking the stable state of the time point as the state mode of the time point.
As shown in fig. 1, it is a flowchart of a time series data processing method according to an embodiment of the present invention, including the following steps:
In this embodiment of the present invention, the time series data may be sampling data of a timing diagram, which is an arbitrary waveform, and may be periodic sampling or aperiodic sampling, which is not limited in this embodiment of the present invention.
The data in the time series data is sorted in ascending order by time or index.
Wherein E (n) is the fluctuation energy of the current time point, V (n) is the value of the current time point, and V (n-1) is the value of the previous time point.
For the first time point in the time series data, the fluctuation energy thereof may be set to 0, i.e., E (1) ═ 0.
Since the fluctuation energy reflects only the change of the current time point compared to the previous time point, only a relative state, not a final state, the final state of each time point is not only related to the state of the previous time point, but also possibly related to the state of one or more time points after the previous time point. Therefore, in the embodiment of the present invention, the temporary state of each time point may be determined according to the fluctuation energy of the time point, and then the steady state of the time point may be determined in a loop traversal manner according to the absolute value of the fluctuation energy of the time point and the temporary state. For the convenience of description, the absolute value of the fluctuation energy is subsequently denoted as abs (e).
The temporary state represents that the state mode of the current time point is not finally determined and can be influenced by the state mode of one or more time points; the steady state indicates that the state pattern at the time point has stabilized and is not affected by the state change at a later time point. In the embodiment of the present invention, three status patterns at each time point may be respectively defined as: a steady mode, an up mode, a down mode. In addition, in practical application, some applications only need to distinguish whether the trend of data in the sequence is stable, so that the state mode can be divided into a stable mode and a fluctuation mode according to different application requirements. Of course, there may be more partitions with different granularities, and the embodiment of the present invention is not limited thereto.
The following description will be given taking as an example the state mode including the steady mode, the ascending mode, and the descending mode. For convenience of description, the three state modes are respectively represented by 0, 1 and 2.
Accordingly, the transient state and the steady state also include the above three state patterns, respectively, and for convenience of description, the transient state is denoted as St, and the steady state is denoted as S.
In determining the temporal status at each time point, the following principles may be followed:
if the fluctuation energy E (n) >0 at the current time point, determining that the temporary state St (n) at the current time point is an ascending mode and recording as St (n) ═ 1;
if the fluctuation energy E (n) at the current time point is less than 0, determining that the temporary state St (n) at the current time point is a descending mode, and recording the temporary state St (n) as St (n) 2;
if the fluctuation energy e (n) at the current time point is 0, the temporal state st (n) at the current time point is determined to be a stationary mode, and is recorded as st (n) 0.
It should be noted that, for the first time point in the time series data, the temporary state thereof may be set to a steady mode, i.e., St (1) ═ 0.
In determining the steady state at each time point, the following principles may be followed:
if the absolute value abs (e (n)) of the fluctuation energy at the current time point is equal to or greater than the set fluctuation energy threshold e, that is, abs (e (n)) > e, the transient state st (n) at the current time point is taken as the steady state at the current time point, that is, the transient state at the current time point becomes the steady state;
if the absolute value abs (e (n)) of the fluctuation energy at the current time point is smaller than the fluctuation energy threshold e, and the transient state St (n) at the current time point is the same as the transient state St (n-1) at the previous time point, i.e., abs (e (n)) < e and St (n) St (n-1), the last steady state (denoted as Sp) is taken as the steady state at the current time point, that is, the steady state at the current time point inherits the last steady state Sp; it should be noted that the last stable state refers to a stable state that is the latest before the current time point, but not the stable state at the previous time point;
if the absolute value abs (E (n)) of the fluctuation energy at the current time point is smaller than the fluctuation energy threshold e, and the temporal state St (n) at the current time point is different from the temporal state St (n-1) at the previous time point, that is, abs (E (n)) < e and St (n)! St (n-1), unsteady state tracking is performed using a dynamic observation window, and the steady state at each time point within the observation window is determined.
The process of performing unsteady state tracking using a dynamic observation window will be described in detail later.
As mentioned above, the steady state indicates that the state pattern at the time point has stabilized and is not affected by the state change at the later time point, and therefore, after the steady state at each time point is determined, the steady state at each time point can be output. Of course, the stable state at each time point may also be determined, that is, the state pattern at the time point is output, which is not limited in this embodiment of the present invention.
Different status patterns can be indicated by different labels, such as the aforementioned numbers 0, 1, 2 for the steady mode, the rising mode, and the falling mode, respectively. Accordingly, at the time of output, the state pattern corresponding to each time point may be output in the form of a state pattern sequence or a table or the like. Further, in order to make the change of the data more intuitive, the status patterns of the time points in the time series data may also be presented in the form of a waveform diagram, wherein different status patterns may be presented in different representations, such as different status patterns representing data points by using different colors or shapes, etc.
As shown in fig. 2, it is a flowchart of performing unsteady state tracking by using a dynamic observation window in the embodiment of the present invention, and the flowchart includes the following steps:
The dynamic observation window means that the observation window is dynamically changed, namely the observation window sequentially extends backwards from the current time point of opening the dynamic observation window, and a subsequent time point is added each time. And each subsequent time point within the observation window is added, the following step 202 needs to be executed again.
Due to the uncertainty of the variation of the time series data, different termination conditions can be set for different variation characteristics. In the embodiment of the present invention, the termination condition may include any one of:
(1) a specific time point appears in the observation window, wherein the specific time point refers to a time point of which the absolute value abs (E (i)) of the fluctuation energy is greater than or equal to the fluctuation energy threshold;
(2) the observation window reaches a set length threshold value w;
(3) the absolute value abs (sum (e)) of the accumulated energy at all time points in the observation window is equal to or greater than the fluctuation energy threshold.
That is, the observation window is terminated as soon as any one of the above conditions is satisfied.
And step 204, terminating the observation window, and determining the stable state of each time point in the observation window according to the termination condition.
Because different change characteristics of the time points in the observation window are met under different termination conditions, the stable state of the time points in the observation window also needs to be determined according to the characteristics of the time points so as to ensure the accuracy of the finally determined state mode of each time point.
Corresponding to the above termination conditions, there may be specifically the following cases:
1) in the event that the specific point in time within the observation window occurs to terminate the observation window: moving the termination point of the observation window to the previous time point of the specific time point to obtain an updated observation window; if the length of the updated observation window is greater than half of the set length threshold value w, setting the stable state of each time point in the updated observation window as the last stable state Sp, otherwise, setting the stable state of each time point in the updated observation window as a stable mode;
2) in case the observation window reaches a set length threshold w terminating the observation window: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is larger than the fluctuation energy threshold, that is, abs (sum (e)) > e, and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is larger than the fluctuation energy threshold, that is, abs (sum (e)) > e, and the accumulated energy sum (e) < 0; setting a steady state at each time point within the observation window to a steady mode if an absolute value abs (sum (e)) of the accumulated energy is equal to or less than the fluctuation energy threshold, i.e., abs (sum (e)) < ═ e;
3) in a case where the absolute value abs (sum (e)) of the accumulated energy at all time points within the observation window is equal to or greater than the fluctuation energy threshold value, the observation window is terminated: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is larger than the fluctuation energy threshold, that is, abs (sum (e)) > e, and the accumulated energy sum (e)) is > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0.
According to the time series data processing method provided by the embodiment of the invention, the fluctuation energy of each time point is calculated and calculated aiming at the time series data, and the state mode of each time point is determined by circularly traversing each time point in the time series by using the fluctuation energy of each time point. Further, the division of a stable state and a non-stable state is proposed, and in the traversal process, the temporary state of each time point is determined according to the fluctuation energy of each time point; and then determining the stable state of the time point according to the absolute value and the temporary state of the fluctuation energy of the time point, so as to obtain the state mode of each time point.
Further, when the fluctuation energy of the current time point is small, if the transient state of the current time point is different from the transient state of the previous time point, the dynamic observation window is used for performing unsteady state tracking to determine the steady state of each time point in the observation window, that is, in the case that the state pattern of the current time point cannot be determined only according to the fluctuation energy of the current time point and the transient state of the previous time point, the state pattern of the current time point is finally determined by considering the transient state conditions of one or more time points after the current time point, so that the determination result of the state pattern of each time point can be more accurate, the continuation of the transient state can be accurately and effectively divided, and the improper division of the state pattern with long time persistence due to the transient change of the energy direction can be reduced. In addition, the influence of energy accumulation change with a certain length or time on the current time point can be accurately judged through unsteady state tracking, and the accuracy of state mode judgment is improved.
Correspondingly, an embodiment of the present invention further provides a time series data processing apparatus, as shown in fig. 3, which is a structural block diagram of the apparatus.
In this embodiment, the apparatus includes the following modules:
a data obtaining module 301, configured to obtain time series data, where each data in the time series is a value of a time point; for example, the time sequence map is periodically or non-periodically sampled to obtain time sequence data, or time-based sampling data provided by a user, etc.;
a fluctuation energy calculation module 302, configured to sequentially calculate fluctuation energy e (n) ═ V (n) — V (n-1) at each time point, where e (n) is the fluctuation energy at the current time point, V (n) is the value at the current time point, and V (n-1) is the value at the previous time point;
a traversing module 303, configured to cycle through each time point in the time sequence, and determine a state of each time point according to the fluctuation energy of each time point, where the state includes: transient state and steady state;
an output module 304, configured to use the stable state at the time point as the state mode at the time point; the status patterns include: a steady mode, an up mode, a down mode.
Since the fluctuation energy reflects only the change of the current time point compared to the previous time point, only a relative state, not a final state, the final state of each time point is not only related to the state of the previous time point, but also possibly related to the state of one or more time points after the previous time point. Therefore, in the embodiment of the present invention, the temporary state of each time point may be determined according to the fluctuation energy of the time point, and then the steady state of the time point may be determined in a loop traversal manner according to the absolute value of the fluctuation energy of the time point and the temporary state. Accordingly, the traversing module may include: a transient state determination module and a stable state determination module; wherein:
the temporary state determining module is used for circularly traversing each time point in the time sequence and determining the temporary state of the time point according to the fluctuation energy of the time point;
and the steady state determining module is used for determining the steady state of the time point according to the absolute value of the fluctuation energy of the time point and the temporary state.
The temporary state determining module may specifically determine the temporary state at each time point according to the following principle: when the fluctuation energy E (n) >0 at the current time point, determining that the temporary state St (n) at the current time point is an ascending mode; when the fluctuation energy E (n) at the current time point is less than 0, determining that the temporary state St (n) at the current time point is a descending mode; when the fluctuation energy e (n) at the current time point is 0, the temporal state st (n) at the current time point is determined to be the stationary mode. In addition, for the first time point in the time-series data, its temporal state may be set to a stationary mode, i.e., St (1) ═ 0.
The steady state determination module may specifically include the following units:
a determination unit for determining whether or not the absolute value abs (e (n)) of the fluctuation energy at the current time point is equal to or greater than a set fluctuation energy threshold, and whether or not the provisional state St (n) at the current time point is the same as the provisional state St (n-1) at the previous time point;
a first steady-state determination unit configured to take a temporary state st (n) at a current time point as a steady state at the current time point when an absolute value abs (e (n)) of fluctuation energy at the current time point is equal to or greater than a set fluctuation energy threshold;
a second steady-state determination unit configured to take the last steady state as the steady state at the current time point when the absolute value abs (e (n)) of the fluctuation energy at the current time point is smaller than the fluctuation energy threshold and the transient state St (n) at the current time point is the same as the transient state St (n-1) at the previous time point;
and a third steady-state determination unit configured to perform unsteady-state tracking using a dynamic observation window and determine a steady state at each time point within the observation window when an absolute value abs (e (n)) of fluctuation energy at a current time point is smaller than the fluctuation energy threshold and a transient state St (n) at the current time point is different from a transient state St (n-1) at a previous time point.
The third steady-state determining unit is specifically configured to open a dynamic observation window, record the length of the observation window, calculate the accumulated energy sum (e) and the absolute value abs (sum (e)) of the accumulated energy at all time points in the observation window, terminate the observation window after the observation window meets a termination condition, and determine a steady state of each time point in the observation window according to the termination condition.
The termination condition and the manner of determining the stable state of each time point in the observation window according to the termination condition may be referred to the description in the foregoing embodiment of the method of the present invention, and are not described herein again.
According to the time series data processing device provided by the embodiment of the invention, the fluctuation energy of each time point is calculated and calculated aiming at the time series data, and the state mode of each time point is determined by circularly traversing each time point in the time series by using the fluctuation energy of each time point. Further, the division of a stable state and a non-stable state is proposed, and in the traversal process, the temporary state of each time point is determined according to the fluctuation energy of each time point; and then determining the stable state of the time point according to the absolute value and the temporary state of the fluctuation energy of the time point, so as to obtain the state mode of each time point.
Further, when the fluctuation energy of the current time point is small, if the transient state of the current time point is different from the transient state of the previous time point, the dynamic observation window is used for performing unsteady state tracking to determine the steady state of each time point in the observation window, that is, in the case that the state pattern of the current time point cannot be determined only according to the fluctuation energy of the current time point and the transient state of the previous time point, the state pattern of the current time point is finally determined by considering the transient state conditions of one or more time points after the current time point, so that the determination result of the state pattern of each time point can be more accurate, the continuation of the transient state can be accurately and effectively divided, and the improper division of the state pattern with long time persistence due to the transient change of the energy direction can be reduced. In addition, the influence of energy accumulation change with a certain length or time on the current time point can be accurately judged through unsteady state tracking, and the accuracy of state mode judgment is improved.
Fig. 4 is a block diagram showing another structure of the time-series data processing apparatus according to the embodiment of the present invention.
Unlike the embodiment shown in fig. 3, in this embodiment, the apparatus further includes:
a display module 401, configured to display a status pattern of each time point in the time series data.
Different status patterns can be indicated by different labels, such as the aforementioned numbers 0, 1, 2 for the steady mode, the rising mode, and the falling mode, respectively. Accordingly, when the output module 304 outputs the state pattern at each time point, the state pattern corresponding to each time point may be output in the form of a state pattern sequence or a table.
Further, the presentation module 401 may present the status patterns of the time points in the time series data in a waveform diagram manner, where different status patterns may be presented in different representations, such as different status patterns representing data points by using different colors or shapes, etc.
It should be noted that, for each embodiment of the above time series data processing apparatus, since the function implementation of each module and unit is similar to that in the corresponding method, the description of each embodiment of the apparatus is relatively simple, and relevant parts can be referred to the description of the corresponding parts of the method embodiment.
By utilizing the time sequence data processing method and the time sequence data processing device provided by the embodiment of the invention, the segmentation and state mode judgment of the time sequence data of any waveform, such as an electrocardiogram, an electroencephalogram, a current-voltage signal in production and manufacturing, a K line of stock transaction, a time domain waveform of a voice signal and the like, can be realized, and the time sequence data processing method and the time sequence data processing device are not influenced by noise data, do not need training data and have higher accuracy and universality. According to the state mode of each time point in the time sequence data, the change point of the time sequence can be obtained, or different state modes can be combined into a composite mode, so that effective information is provided for industry analysis and application.
Compared with the prior art, the scheme of the invention can accurately identify the state mode of the steady-state or non-steady-state time sequence data which is sampled at will, does not need data distribution hypothesis, has or does not have a time sequence period, not only can realize off-line identification, but also can realize on-line identification, has strong universality and is not limited by application environment.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. Furthermore, the above-described system embodiments are merely illustrative, wherein modules and units illustrated as separate components may or may not be physically separate, i.e., may be located on one network element, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Those skilled in the art will appreciate that all or part of the steps in the above method embodiments may be implemented by a program to instruct relevant hardware to perform the steps, and the program may be stored in a computer-readable storage medium, referred to herein as a storage medium, such as: ROM/RAM, magnetic disk, optical disk, etc.
Correspondingly, the embodiment of the invention also provides a device for the time series data processing method, and the device is an electronic device, such as a mobile terminal, a computer, a tablet device, a medical device, a fitness device, a personal digital assistant and the like. The electronic device may include one or more processors, memory; wherein the memory is used for storing computer executable instructions and the processor is used for executing the computer executable instructions to realize the method of the previous embodiments.
The present invention has been described in detail with reference to the embodiments, and the description of the embodiments is provided to facilitate the understanding of the method and apparatus of the present invention, and is intended to be a part of the embodiments of the present invention rather than the whole embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without any creative effort shall fall within the protection scope of the present invention, and the content of the present description shall not be construed as limiting the present invention. Therefore, any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (14)
1. A method for processing time-series data, the method comprising:
sampling a time sequence diagram to be processed to obtain time sequence data, wherein each data in the time sequence is a value of a time point, and the time sequence diagram is any one of the following: electrocardiogram, electroencephalogram, current-voltage signal in production and manufacture, K line of stock exchange, and time domain waveform of voice signal;
calculating the fluctuation energy E (n) = V (n) — V (n-1) of each time point in sequence, wherein E (n) is the fluctuation energy of the current time point, V (n) is the value of the current time point, and V (n-1) is the value of the previous time point;
circularly traversing each time point in the time sequence, and determining the state of each time point according to the fluctuation energy of each time point, wherein the state comprises the following steps: transient state and steady state;
taking the stable state of the time point as the state mode of the time point; the status patterns include: a steady mode, an ascending mode, a descending mode;
outputting a time series change point according to the state mode of each time point in the time series data;
the step of circularly traversing each time point in the time sequence, and the step of determining the state of each time point according to the fluctuation energy of each time point comprises the following steps:
circularly traversing each time point in the time sequence, and determining the temporary state of the time point according to the fluctuation energy of the time point;
determining the stable state of the time point according to the absolute value and the temporary state of the fluctuation energy of the time point, specifically comprising:
if the absolute value abs (E (n)) of the fluctuation energy at the current time point is greater than or equal to the set fluctuation energy threshold, taking the temporary state St (n) at the current time point as the stable state at the current time point;
if the absolute value abs (e (n)) of the fluctuation energy at the present time point is smaller than the fluctuation energy threshold, and the transient state St (n) at the present time point is the same as the transient state St (n-1) at the previous time point, the last steady state is taken as the steady state at the present time point;
if the absolute value abs (E (n)) of the fluctuation energy at the current time point is smaller than the fluctuation energy threshold and the temporal state St (n) at the current time point is different from the temporal state St (n-1) at the previous time point, performing unsteady state tracking using a dynamic observation window, and determining the steady state at each time point within the observation window.
2. The method of claim 1, wherein determining the temporal state of the time point from the fluctuating energy of the time point comprises:
if the fluctuation energy E (n) >0 at the current time point, determining that the temporary state St (n) at the current time point is in the ascending mode;
determining that the temporary state St (n) of the current time point is a descending mode if the fluctuation energy E (n) of the current time point is less than 0;
if the fluctuation energy e (n) =0 at the current time point, it is determined that the temporal state st (n) at the current time point is the stationary mode.
3. The method of claim 1, wherein the non-steady state tracking using the dynamic observation window, and determining the steady state at the current time point comprises:
opening a dynamic observation window;
recording the length of the observation window and calculating the accumulated energy sum (E) and the absolute value abs (sum (E)) of the accumulated energy at all time points within the observation window;
judging whether the observation window meets a termination condition;
if yes, terminating the observation window, and determining the stable state of each time point in the observation window according to the termination condition.
4. The method according to claim 3, wherein the termination condition comprises any one of:
a specific time point appears in the observation window, wherein the specific time point refers to a time point of which the absolute value abs (E (i)) of the fluctuation energy is greater than or equal to the fluctuation energy threshold;
the observation window reaches a set length threshold value w;
the absolute value abs (sum (E)) of the accumulated energy of all time points in the observation window is greater than or equal to the fluctuation energy threshold;
the determining the stable state of each time point in the observation window according to the termination condition includes:
in the event that the specific point in time within the observation window occurs to terminate the observation window: moving the termination point of the observation window to the previous time point of the specific time point to obtain an updated observation window; if the length of the updated observation window is greater than half of the set length threshold value w, setting the stable state of each time point in the updated observation window as the last stable state, otherwise, setting the stable state of each time point in the updated observation window as a stable mode;
in case the observation window reaches a set length threshold w terminating the observation window: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0; setting a steady state at each time point within the observation window to a steady mode if an absolute value abs (sum (e)) of the accumulated energy is equal to or less than the fluctuation energy threshold;
in a case where the absolute value abs (sum (e)) of the accumulated energy at all time points within the observation window is equal to or greater than the fluctuation energy threshold value, the observation window is terminated: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0.
5. The method of claim 1, wherein sampling the timing graph to be processed comprises:
the timing diagrams to be processed are periodically sampled or the timing diagrams to be processed are non-periodically sampled.
6. The method according to any one of claims 1 to 5, further comprising:
and displaying the state mode of each time point in the time sequence data.
7. A time-series data processing apparatus, characterized in that the apparatus comprises:
the data acquisition module is configured to sample a time sequence diagram to obtain time sequence data, where each data in the time sequence is a value of a time point, and the time sequence diagram is any one of the following: electrocardiogram, electroencephalogram, current-voltage signal in production and manufacture, K line of stock exchange, and time domain waveform of voice signal;
the fluctuation energy calculation module is used for calculating the fluctuation energy E (n) = V (n) -V (n-1) of each time point in sequence, wherein E (n) is the fluctuation energy of the current time point, V (n) is the value of the current time point, and V (n-1) is the value of the previous time point;
a traversing module, configured to cycle through each time point in the time sequence, and determine a state of each time point according to fluctuation energy of each time point, where the state includes: transient state and steady state;
the output module is used for taking the stable state of the time point as the state mode of the time point and outputting the change point of the time sequence according to the state mode of each time point in the time sequence data; the status patterns include: a steady mode, an ascending mode, a descending mode;
the traversal module comprises:
the temporary state determining module is used for circularly traversing each time point in the time sequence and determining the temporary state of the time point according to the fluctuation energy of the time point;
the steady state determining module is used for determining the steady state of the time point according to the absolute value of the fluctuation energy of the time point and the temporary state; the steady state determination module comprises:
a determination unit for determining whether or not the absolute value abs (e (n)) of the fluctuation energy at the current time point is equal to or greater than a set fluctuation energy threshold, and whether or not the provisional state St (n) at the current time point is the same as the provisional state St (n-1) at the previous time point;
a first steady-state determination unit configured to take a temporary state st (n) at a current time point as a steady state at the current time point when an absolute value abs (e (n)) of fluctuation energy at the current time point is equal to or greater than a set fluctuation energy threshold;
a second steady-state determination unit configured to take the last steady state as the steady state at the current time point when the absolute value abs (e (n)) of the fluctuation energy at the current time point is smaller than the fluctuation energy threshold and the transient state St (n) at the current time point is the same as the transient state St (n-1) at the previous time point;
and a third steady-state determination unit configured to perform unsteady-state tracking using a dynamic observation window and determine a steady state at each time point within the observation window when an absolute value abs (e (n)) of fluctuation energy at a current time point is smaller than the fluctuation energy threshold and a transient state St (n) at the current time point is different from a transient state St (n-1) at a previous time point.
8. The apparatus of claim 7,
the temporary state determination module is specifically configured to determine that the temporary state st (n) at the current time point is an ascending mode when the fluctuation energy e (n) at the current time point is > 0; when the fluctuation energy E (n) at the current time point is less than 0, determining that the temporary state St (n) at the current time point is a descending mode; when the fluctuation energy e (n) =0 at the current time point, it is determined that the transient state st (n) at the current time point is the stationary mode.
9. The apparatus of claim 7,
the third steady-state determining unit is specifically configured to open a dynamic observation window, record the length of the observation window, calculate the accumulated energy sum (e) and the absolute value abs (sum (e)) of the accumulated energy at all time points in the observation window, terminate the observation window after the observation window meets a termination condition, and determine a steady state of each time point in the observation window according to the termination condition.
10. The apparatus of claim 9, wherein the termination condition comprises any one of:
a specific time point appears in the observation window, wherein the specific time point refers to a time point of which the absolute value abs (E (i)) of the fluctuation energy is greater than or equal to the fluctuation energy threshold;
the observation window reaches a set length threshold value w;
the absolute value abs (sum (E)) of the accumulated energy of all time points in the observation window is greater than or equal to the fluctuation energy threshold;
the third steady-state determination unit determines the steady state at each time point within the observation window in the following manner:
in the event that the specific point in time within the observation window occurs to terminate the observation window: moving the termination point of the observation window to the previous time point of the specific time point to obtain an updated observation window; if the length of the updated observation window is greater than half of the set length threshold value w, setting the stable state of each time point in the updated observation window as the last stable state, otherwise, setting the stable state of each time point in the updated observation window as a stable mode;
in case the observation window reaches a set length threshold w terminating the observation window: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0; setting a steady state at each time point within the observation window to a steady mode if an absolute value abs (sum (e)) of the accumulated energy is equal to or less than the fluctuation energy threshold;
in a case where the absolute value abs (sum (e)) of the accumulated energy at all time points within the observation window is equal to or greater than the fluctuation energy threshold value, the observation window is terminated: setting a steady state at each time point within the observation window to an ascending mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) > 0; setting a steady state at each time point within the observation window to a falling mode if an absolute value abs (sum (e)) of the accumulated energy is greater than the fluctuation energy threshold and the accumulated energy sum (e) < 0.
11. The apparatus of claim 7, wherein the data acquisition module is to sample a timing graph comprising: the timing diagram is sampled periodically or the timing diagram is sampled aperiodically.
12. The apparatus of any one of claims 7 to 11, further comprising:
and the display module is used for displaying the state mode of each time point in the time sequence data.
13. An electronic device, comprising: one or more processors, memory;
the memory is for storing computer-executable instructions, and the processor is for executing the computer-executable instructions to implement the method of any one of claims 1 to 6.
14. A readable storage medium having stored thereon instructions that are executed to implement the method of any of claims 1 to 6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910526163.6A CN110232132B (en) | 2019-06-18 | 2019-06-18 | Time series data processing method and device |
| PCT/CN2019/095065 WO2020252822A1 (en) | 2019-06-18 | 2019-07-08 | Time series data processing method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910526163.6A CN110232132B (en) | 2019-06-18 | 2019-06-18 | Time series data processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110232132A CN110232132A (en) | 2019-09-13 |
| CN110232132B true CN110232132B (en) | 2020-11-06 |
Family
ID=67859606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910526163.6A Active CN110232132B (en) | 2019-06-18 | 2019-06-18 | Time series data processing method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110232132B (en) |
| WO (1) | WO2020252822A1 (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103690156A (en) * | 2013-11-22 | 2014-04-02 | 东软熙康健康科技有限公司 | Heart rate acquiring method and electrocardiosignal processing method |
| CN103840970A (en) * | 2014-01-24 | 2014-06-04 | 珠海多玩信息技术有限公司 | Method and device for obtaining running status of service |
| CN104573321A (en) * | 2014-12-11 | 2015-04-29 | 国家电网公司 | Recognition and processing method of bad data of dissolved gas in transformer oil |
| US9110111B1 (en) * | 2013-04-26 | 2015-08-18 | Texas Instruments Incorporated | Methods and systems to determine a final value of random telegraph noise time constant and magnitude |
| CN105607475A (en) * | 2015-12-24 | 2016-05-25 | 西安交通大学 | Electrical appliance operating state recognition method based on dual-sliding window |
| CN108804703A (en) * | 2018-06-19 | 2018-11-13 | 北京焦点新干线信息技术有限公司 | A kind of data exception detection method and device |
| CN108959704A (en) * | 2018-05-28 | 2018-12-07 | 华北电力大学 | A kind of rewards and punishments weight type simulation sequence similarity analysis method considering metamorphosis |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2284769B1 (en) * | 2009-07-16 | 2013-01-02 | European Space Agency | Method and apparatus for analyzing time series data |
| CN104268660A (en) * | 2014-10-13 | 2015-01-07 | 国家电网公司 | Trend recognition method for electric power system predication-like data |
| JP6794919B2 (en) * | 2017-04-28 | 2020-12-02 | 横河電機株式会社 | Process control system and data processing method |
-
2019
- 2019-06-18 CN CN201910526163.6A patent/CN110232132B/en active Active
- 2019-07-08 WO PCT/CN2019/095065 patent/WO2020252822A1/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9110111B1 (en) * | 2013-04-26 | 2015-08-18 | Texas Instruments Incorporated | Methods and systems to determine a final value of random telegraph noise time constant and magnitude |
| CN103690156A (en) * | 2013-11-22 | 2014-04-02 | 东软熙康健康科技有限公司 | Heart rate acquiring method and electrocardiosignal processing method |
| CN103840970A (en) * | 2014-01-24 | 2014-06-04 | 珠海多玩信息技术有限公司 | Method and device for obtaining running status of service |
| CN104573321A (en) * | 2014-12-11 | 2015-04-29 | 国家电网公司 | Recognition and processing method of bad data of dissolved gas in transformer oil |
| CN105607475A (en) * | 2015-12-24 | 2016-05-25 | 西安交通大学 | Electrical appliance operating state recognition method based on dual-sliding window |
| CN108959704A (en) * | 2018-05-28 | 2018-12-07 | 华北电力大学 | A kind of rewards and punishments weight type simulation sequence similarity analysis method considering metamorphosis |
| CN108804703A (en) * | 2018-06-19 | 2018-11-13 | 北京焦点新干线信息技术有限公司 | A kind of data exception detection method and device |
Non-Patent Citations (1)
| Title |
|---|
| 《基于同步多维数据流的脑网络动态特征辨识方法研究》;马洒洒等;《计算机应用研究》;20171130;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110232132A (en) | 2019-09-13 |
| WO2020252822A1 (en) | 2020-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110851338B (en) | Abnormality detection method, electronic device, and storage medium | |
| US9612791B2 (en) | Method, system and storage medium for monitoring audio streaming media | |
| CN109697207A (en) | The abnormality monitoring method and system of time series data | |
| CN108959004B (en) | Disk failure prediction method, device, equipment and computer readable storage medium | |
| CN113518011A (en) | Abnormality detection method and apparatus, electronic device, and computer-readable storage medium | |
| CN112435142B (en) | Power load identification method and load power utilization facility knowledge base construction method thereof | |
| CN113268403A (en) | Time series analysis and prediction method, device, equipment and storage medium | |
| CN105429792A (en) | User behavior flow obtaining method and device and user behavior analysis method and system | |
| CN110232132B (en) | Time series data processing method and device | |
| CN114757229A (en) | Signal processing method, signal processing device, electronic apparatus, and medium | |
| CN107894837A (en) | Dynamic sentiment analysis model sample processing method and processing device | |
| CN104706366A (en) | Distraction testing method, device and system | |
| CN109009051A (en) | Method for measuring heart rate, device, equipment and storage medium | |
| CN110198299B (en) | Intrusion detection method and device | |
| Minh et al. | Parallel workload modeling with realistic characteristics | |
| CN107688619A (en) | A kind of daily record data processing method and processing device | |
| CN112783937B (en) | Product production monitoring method and device based on big data and cloud computing | |
| CN108956160B (en) | Method, system and device for measuring warning sound parameters of vehicle-mounted entertainment equipment and computer readable storage medium | |
| Yu et al. | Network inference and change point detection for piecewise-stationary time series | |
| CN114533010A (en) | Heart rate detection method and device | |
| Upham et al. | Limits on the application of statistical correlations to continuous response data | |
| CN112308099A (en) | Sample feature importance determination method, and classification model training method and device | |
| CN109008993A (en) | A kind of vital sign data collection control method and device | |
| CN109885467A (en) | Alarm method, device, storage medium and the electronic equipment of data fluctuations | |
| Kim et al. | Evaluating the regularity of human behavior from mobile phone usage logs |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |