CN110956675B - Method and device for automatically generating technology maturity curve - Google Patents
Method and device for automatically generating technology maturity curve Download PDFInfo
- Publication number
- CN110956675B CN110956675B CN201911031409.9A CN201911031409A CN110956675B CN 110956675 B CN110956675 B CN 110956675B CN 201911031409 A CN201911031409 A CN 201911031409A CN 110956675 B CN110956675 B CN 110956675B
- Authority
- CN
- China
- Prior art keywords
- curve
- sequence
- maturity
- formula
- fitting
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/203—Drawing of straight lines or curves
Abstract
The invention discloses a method and a device for automatically generating a technical maturity curve, wherein the method comprises the following steps: acquiring a literature sequence and a patent sequence; acquiring local bandwidth of a kernel density function according to the document sequence, estimating adaptive kernel density of the document sequence according to the local bandwidth, and identifying an inverted bell-shaped characteristic to generate a foam period curve; preprocessing the patent sequence, and performing curve fitting on the patent sequence to generate a maturation period curve; and generating a final maturity curve according to the foam period curve and the maturity curve. The method can analyze the maturity of the technology without manual operation and expert opinions, has a certain reference value for the state, enterprises and scientific researchers to judge the development condition of the technology, and is strong in applicability, simple and easy to implement.
Description
Technical Field
The invention relates to the technical field of computer network information, in particular to a method and a device for automatically generating a technical maturity curve.
Background
The maturity of the technology is often an important consideration in planning and decision making, and is also a commonly used research index in academic research and reports. For example, many enterprises can understand the development of a new technology by considering its maturity in the decision of whether to adopt it. As a result of research, the most common technical maturity model is the technical maturity curve of Gaodena, which describes the development process of an emerging technology from birth to foam stage, through valley, and finally to the mature stable stage, and provides the relative maturity index of the technology in the market or field, and the attitude of people to the technology. It is widely used in enterprise decision-making because it can visually represent the development of various emerging technologies and the risks of new technologies. The administrative and administrative staff will utilize the annual reports of the technology maturity curves of high-tech companies to find trends in the development of emerging technologies, avoiding the excessive expectations and subsequent investments in technology due to its high visibility between media and users. At present, methods for generating technical maturity curves rely mainly on manual analysis. The internal team of Gaodena deduces the mature evolution speed of various new technologies and the time required for reaching the maturity through professional analysis and prediction.
Much work has been directed to finding measures of technical maturity through specific analysis and experimentation, most of which are based on the definitions in official reports: the maturity curve can be divided into a superposition of two curves: the first curve is an inverted bell-shaped curve representing the initial heat change, and the second curve is an S-shaped curve representing the new technology gradually stabilizing and putting into actual production. However, the official reports of Gaodena corporation indicate that in practice there is not just one foam stage in the development cycle for some technologies that will continue to cycle between high heat and bottom valleys, and these technologies are referred to as "Phoenix-type technologies".
In the related art, a mathematical method for simulating a technical maturity curve is proposed, which mainly comprises the step of dividing news articles into two types according to the heading semantics, and respectively dividing the two stages for representing the maturity curve. And performing Sigmoid curve fitting, standardized merging and high-order polynomial function fitting on the data to finally obtain a technical maturity curve. However, this method can only fit an inverted bell-shaped curve and so can only be used for techniques with only one foam period. Similarly, in the related art, two parts of the technical maturity curve are respectively expressed by using papers and patent numbers, and then the two parts are respectively approximated by using a Sigmoid function, and then are combined. But equally only inapplicable to "phoenix-type technology".
In addition to using literature statistical data, there are also methods that exploit semantics. For example, the basic idea is that the large number of documents cannot fully represent the enthusiasm of the technology, so its method is to classify the content of articles according to their emotional colors, in addition to the number of articles. However, this method involves manual operations and is not fully automated.
In summary, regarding the assessment of the technical maturity, existing research generally does not provide an automatic method suitable for general technology, and there are manual operation steps, or the robustness is low, or only suitable for a certain situation, the applicability is poor, and needs to be solved.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, one purpose of the invention is to provide an automatic generation method of a technology maturity curve, which can analyze the technology maturity without manual operation and expert opinions, has a certain reference value for the state, enterprise and scientific research workers to judge the development condition of the technology, and has strong applicability and is simple and easy to implement.
The invention also aims to provide a device for automatically generating the technical maturity curve.
In order to achieve the above object, an embodiment of the invention provides an automatic generation method of a technical maturity curve, which includes the following steps: acquiring a literature sequence and a patent sequence; acquiring local bandwidth of a kernel density function according to the literature sequence, estimating adaptive kernel density of the literature sequence according to the local bandwidth, and identifying an inverted bell-shaped characteristic to generate a foam period curve; preprocessing the patent sequence, and performing curve fitting on the patent sequence to generate a maturity curve; and generating a final maturity curve according to the foam period curve and the maturity curve.
According to the automatic generation method of the technical maturity curve, historical data of papers and patents related to the technology are used, the curve is simulated through nuclear density estimation and curve fitting, and the technical maturity curve is obtained through synthesis, so that the technical maturity can be analyzed without manual operation and expert opinions, certain reference value is provided for the state, enterprises and scientific researchers to judge the development condition of the technology, and the method is high in applicability and simple and easy to implement.
In addition, the automatic generation method of the technical maturity curve according to the above embodiment of the present invention may further have the following additional technical features:
further, in an embodiment of the present invention, the definition formula of the local bandwidth is:
wherein the data point xiL (x) is a literature sequence, x ∈ { x ∈ }1，x2，...，xnS is a smoothing factor.
Further, in an embodiment of the present invention, the adaptive kernel density estimation formula is:
where K (x) is a kernel function.
Further, in an embodiment of the present invention, the preprocessing the patent sequence numbers and curve fitting the patent sequences include: and accumulating the patent sequence to obtain an accumulated patent sequence, wherein the operation formula is as follows:
wherein, p (x) is a patent sequence;
performing data normalization on the accumulated patent sequence, wherein the normalization formula is as follows:
curve fitting the normalized patent sequence using Gompertz function, wherein the fitting formula is:
where a, b, c are parameters of the curve, a is the asymptote, b controls the displacement of the curve on the x-axis, and c controls the growth rate.
Further, in one embodiment of the present invention, the final maturity curve is:
wherein the content of the first and second substances,
m is a numerical point.
In order to achieve the above object, an embodiment of another aspect of the present invention provides an apparatus for automatically generating a technical maturity curve, including: the acquisition module is used for acquiring a document sequence and a patent sequence; the foam period curve generation module is used for acquiring the local bandwidth of a kernel density function according to the document sequence, estimating the self-adaptive kernel density of the document sequence according to the local bandwidth and identifying the inverse bell-shaped characteristic so as to generate a foam period curve; a maturity curve generating module, configured to pre-process the patent sequence and perform curve fitting on the patent sequence to generate a maturity curve; and the maturity curve generation module is used for generating a final maturity curve according to the foam period curve and the maturity curve.
According to the automatic generation device of the technical maturity curve, the historical data of papers and patents related to the technology are used, the curve is simulated through nuclear density estimation and curve fitting, and the technical maturity curve is obtained through synthesis, so that the technical maturity can be analyzed without manual operation and expert opinions, certain reference value is provided for the state, enterprises and scientific researchers to judge the development condition of the technology, and the automatic generation device is high in applicability and easy to achieve.
In addition, the automatic generation device of the technical maturity curve according to the above embodiment of the present invention may further have the following additional technical features:
further, in an embodiment of the present invention, the definition formula of the local bandwidth is:
wherein the data point xiL (x) is a literature sequence, x ∈ { x ∈ }1，x2，...，xnS is a smoothing factor.
Further, in an embodiment of the present invention, the adaptive kernel density estimation formula is:
where K (x) is a kernel function.
Further, in an embodiment of the present invention, the maturity curve generating module is further configured to perform an operation of accumulating the patent sequences to obtain an accumulated patent sequence, where the operation formula is:
wherein, p (x) is a patent sequence;
performing data normalization on the accumulated patent sequence, wherein the normalization formula is as follows:
curve fitting the normalized patent sequence using Gompertz function, wherein the fitting formula is:
where a, b, c are parameters of the curve, a is the asymptote, b controls the displacement of the curve on the x-axis, and c controls the growth rate.
Further, in one embodiment of the present invention, the final maturity curve is:
wherein the content of the first and second substances,
m is a numerical point.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flow diagram of a method for automatic generation of a technology maturity curve in accordance with one embodiment of the present invention;
FIG. 2 is a flow chart of a method for automatically generating a technology maturity curve according to an embodiment of the present invention;
FIG. 3 is an example of paper and patent data for a technique according to an embodiment of the present invention;
FIG. 4 is an example of computing a bandwidth satisfying a smoothing condition according to an embodiment of the present invention;
FIG. 5 is an example of adaptive kernel density estimation according to an embodiment of the present invention;
FIG. 6 is an example of identifying a reverse bell-shaped feature according to an embodiment of the present invention;
FIG. 7 is an example of curve fitting according to an embodiment of the present invention;
FIG. 8 is an example of curve synthesis according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of an apparatus for automatically generating a technical maturity curve according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
The present application is based on the recognition and discovery by the inventors of the following problems:
the prediction of the technical maturity is a very important problem, and people can predict the future development trend according to the maturity and invest time and resources more accurately. The most widely known maturity model is mainly the technical maturity curve of high-german corporation. The maturity model is widely applied to enterprise companies and laboratory research, for example, to help risk assessment in enterprise decision making and to provide references to research directions in laboratories. Currently, methods for generating technology maturity curves can only be analyzed and predicted by the expertise of the high-German corporation team. The embodiment of the invention designs a method and a device for automatically generating a technical maturity curve based on literature measurement, and the basic idea is to use historical data of papers and patents related to the technology, simulate the curve through nuclear density estimation and curve fitting respectively, and obtain the technical maturity curve through synthesis, so that the maturity of any technology can be analyzed and predicted without expert opinions or manual analysis.
Further, the problem to be solved by the embodiments of the present invention can be formally defined as: inputting a thesis quantity time sequence l (x) and a patent quantity time sequence p (x) in a technical field, wherein l (x)0) For this technical field in x0Number of papers published in the year, p (x)0) For this technical field in x0Number of patents registered in the year. The output is the maturity curve h (x) for this technique.
There are definitions in official reports: the technical maturity curve can be divided into two parts, wherein the first part is only influenced by stir-frying heat and mainly causes the change of heat by media news; while the second part is dominated by the actual engineering production and the commercial performance growth, combining these two curves is the technology maturity curve. The method of the embodiment of the invention is based on the assumption that the two parts of the technical maturity curve are respectively defined as follows:
foam phase curve: the curved shape of the first portion, such as an inverted bell shape, may be represented by a temporal variation in the number of articles, media articles or related documents of the technology, representing the visibility of the technology during the period of time at the media and at the public.
Maturity curve: the shape of the curve of the second part is an S-shaped curve, which can be represented by the time variation of the patent statistics of the art. Considering that a product of a technology is put into actual production and continuously used after being patented, the influence of the patent is a long-term process in essence, so the accumulated amount of the patent used in the embodiment of the invention in time represents the popularization degree of the technology in actual production.
Therefore, the calculation of the maturity curve is actually the calculation of the curves of the two parts, and the basic idea is to perform kernel density estimation on the literature series and perform curve fitting on the patent series. And finally, synthesizing the results of the two parts to obtain a technical maturity curve, and carrying out the whole calculation under a framework process, wherein the whole process is shown in a figure 1.
The following describes a method and an apparatus for automatically generating a technical maturity curve according to an embodiment of the present invention with reference to the drawings, and first, a method for automatically generating a technical maturity curve according to an embodiment of the present invention will be described with reference to the drawings.
Fig. 2 is a flowchart of a method for automatically generating a technology maturity curve according to an embodiment of the present invention.
As shown in fig. 2, the method for automatically generating the technical maturity curve includes the following steps:
in step S201, a document sequence and a patent sequence are acquired.
It is understood that, as shown in fig. 1, the document number sequence l (x) and the patent number sequence p (x) are input first in the embodiment of the present invention.
In step S202, a local bandwidth of the kernel density function is obtained from the document sequence, and an adaptive kernel density of the document sequence is estimated from the local bandwidth, and an inverse bell-shaped feature is identified to generate a froth phase curve.
It is to be understood that, as shown in fig. 2, calculating the froth phase curve includes: (1) calculating the local bandwidth of the kernel density function; (2) performing adaptive kernel density estimation on the literature sequence; (3) a reverse bell-shaped feature is identified.
Specifically, (1) calculating the local bandwidth of the kernel density function
The time series of literature quantities, l (x), is a broken line, and embodiments of the present invention utilize kernel density estimation to obtain a smooth curve that is shaped as the maturity of the technology. The bandwidth of the kernel density function determines the smoothness of the curve, and has a great influence on the estimation result. To accommodate the inverse bell-shaped feature in the sequence, the local bandwidth is used: a smaller bandwidth is used for the high density part of the document and a larger bandwidth is used for the low density part. In one embodiment of the present invention, the definition formula of the local bandwidth is:
wherein the data point xiL (x) is a literature sequence, x ∈ { x ∈ }1，x2，...，xnS is a smoothing factor.
(2) Adaptive kernel density estimation for document sequences
For the paper sequence l (x), adaptive kernel density estimation is performed using gaussian kernel functions and local bandwidth. And (4) obtaining a curve of a foam stage by weighted accumulation by using Gaussian kernels with different bandwidths at different data points. In one embodiment of the present invention, the adaptive kernel density estimation formula is:
where K (x) is a kernel function.
(3) Identifying inverted bell-shaped features
The portion of the froth phase curve characterized by an inverted bell shape is identified to determine how many froth phases are in the development cycle of the technology.
In step S203, the patent sequence is preprocessed and curve-fitted to generate a maturity curve.
It is understood that, as shown in fig. 2, calculating the maturity curve includes: (1) preprocessing a patent sequence; (2) performing curve fitting on the patent number series; (3) the curves of both parts are combined.
Further, in an embodiment of the present invention, the preprocessing the patent sequence numbers and curve fitting the patent sequences include: and accumulating the patent sequence to obtain an accumulated patent sequence, wherein the operation formula is as follows:
wherein, p (x) is a patent sequence;
and carrying out data standardization on the accumulated patent sequences, wherein the standardization formula is as follows:
curve fitting the normalized patent sequence using Gompertz function, wherein the fitting formula is:
where a, b, c are parameters of the curve, a is the asymptote, b controls the displacement of the curve on the x-axis, and c controls the growth rate.
Specifically, (1) pretreatment of patent sequences
The embodiment of the invention uses the accumulated number of patents to express the maturity period of the technology, so the patent sequence p (x) needs to be accumulated and then can be merged with the literature sequence after data standardization.
(2) Curve fitting of patent series
The time series p (x) of the patent number is also a broken line, and the embodiment of the invention uses a curve fitting method to obtain a smooth curve. For the patent time series, a Gompertz function was used for fitting to obtain a sigmoidal curve of maturity.
(3) Curve combining two parts
The two calculated sections are aligned in time and combined to produce a maturity curve. The maturity curve is obtained by taking the maximum value of the two partial curves, and indicates that the influence of the technology at the time point is dominated by the popular heat or market production.
In step S204, a final maturity curve is generated according to the foam stage curve and the maturation stage curve.
Wherein, in one embodiment of the present invention, the final maturity curve is:
wherein the content of the first and second substances,
m is a numerical point.
The automatic generation method of the technical maturity curve will be described in detail below, taking the techniques "Artificial intelligence (Artificial intelligence)" and "mini computer (Minicomputer)" as examples. Fig. 3 shows data for these two techniques, the solid line is the sequence of the number of papers from 1960 to 2015, and the dotted line is the sequence of the number of patents from 1976 to 2015, as follows:
step 1, calculating local bandwidth of kernel density function
A standard estimate of the nuclear density is:
where k (x) is the kernel function and h is the bandwidth, the degree of smoothing can be controlled. The larger the bandwidth, the flatter the result of the kernel density estimation; conversely, the smaller the bandwidth, the more peaky the result. The basic idea of using adaptive kernel density estimation to better adapt to the inverse bell-shaped features in the literature sequence is to use a smaller bandwidth for the high density part and a larger bandwidth for the low density part, so that the high density part is too steep and the other parts are flat. Therefore, the local core function bandwidth needs to be selected according to the density change of the sequence.
To calculate the bandwidth of the adaptive document sequence, an initial smoothing factor s is first selected0And calculating the local bandwidth. Given a document sequence l (x), x ∈ { x }1，x2，...，xnData point xiThe local bandwidth of (a) is defined as:
where s is a smoothing factor, the larger s, the larger the bandwidth, the smoother the kernel density estimate, and vice versa.
This bandwidth is then used to compute an adaptive kernel density estimate, the specific computation method being described in detail further below. And (4) carrying out derivation on the kernel density estimation result, solving all inflection points, judging whether the smoothing condition is met, if the condition is not met, increasing the smoothing factor by delta s, and repeating the processes until the result meets the condition.
The conditions for judging whether the kernel density estimation is smooth are as follows:
1. the number of inflection points is less than t1A plurality of;
2. the number of numerical points between every two adjacent inflection points is more than t2And (4) respectively.
In the experiment, the initial value s of the smoothing factor0Set to 0.1, increment Δ s to 0.1, t1Is set to 4, t2Set to 3.
For example, for the technique "mini-computer", the local bandwidth is calculated, and after 2 cycles, the condition is satisfied, where the smoothing factor is 0.2, the number of inflection points in the kernel density estimation result is 3, and the derivative and the inflection points in the calculation process are shown in fig. 4.
Local bandwidth h with document sequenceiA smooth curve can be calculated using an adaptive kernel density estimate, using kernel functions of different bandwidths at different data points. The adaptive kernel density estimate is:
the kernel function used in the embodiment of the present invention is a gaussian function:
in addition, the kernel density estimation has an edge effect, i.e. at the edges of the sequence,
it tends to be underestimated because it lacks data outside the boundary. To address this problem, embodiments of the present invention use a method of reflection data, considering the literature sequence l (x), x ∈ { x }1，x2，...，xnThe prevailing density of the sequences is concentrated to the right, so the whole sequence is inverted xn，xn+1＝2xn-xn-1，...，x2n-1＝2xn-x1And is connected to the back of the original sequence, and the new sequence is as follows:
finally, self-adaptive kernel density estimation is carried out, and x is not more than xnThe foam stage curve was obtained.
The adaptive kernel density estimation results for the two examples "artificial intelligence" and "mini-computer" are shown in FIG. 5, with the solid line being the original sequence and the dotted line being the kernel density estimation result.
Step 3, identifying the characteristics of the inverted bell shape
Next, portions of each foam stage are extracted according to the foam stage curve.
The curve is first normalized by dividing the curve by its maximum value so that the curve maximum value is 1. Then calculating the derivative and the inflection point of the curve, and finding out that the curve value is greater than the initial threshold value t3Coordinate z of0Set as the starting point. Then openChecking inflection points one by one:
if the inflection point is smaller than the starting point z0Skipping to continue checking the next inflection point; if the inflection point is transited from the derivative being larger than zero to being smaller than zero, namely the curve is at a local maximum, skipping to continuously check the next inflection point; otherwise, if the last numerical point at the end of the inverse bell shape is incremented by one (if there is no inverse bell shape before, then take the starting point z0) The presence of a curve between the current point of inflection and the current point of inflection exceeds a threshold t4The value points of (a) are determined, then the curve forms an inverted bell-shaped curve. Then the check continues with the next inflection point. If all the inflection points are checked, the end point of the curve is checked finally as long as the curve between the end point of the curve and the last numerical point of the end of the inverse bell shape plus one exceeds the threshold value t4Under the conditions of (3), an inverted bell-shaped curve can be formed.
In the experiment, t3Is set to 0.1, t4Set to 0.5. Taking the technique "artificial intelligence" as an example, the starting point z0For 1981, the inflection points 1988, 1998 and 2012, are all larger than the starting point, so the inflection point 1988 is first examined when the curve is at a local maximum, skipped. Then check 1998 that the curve is at a local minimum and the threshold requirement is met, then 1988 to 1998 are inverted bell-shaped curves. Then check 2012, local maximum, skip. Finally, the curve end point 2015 is checked to meet the requirements, so that a reverse bell shape is formed with the last reverse bell end point 1998 plus one, i.e. 1999. As shown in fig. 6, this technique has two foam stages.
Step 4, preprocessing the patent sequence
After steps 1 to 3 are completed, the calculation of the document sequence is completed. The processing of the patent number is started next. Firstly, the patent sequence p (x) is accumulated, that is:
the accumulated patent sequences were then subjected to data normalization:
After pre-processing, a curve fit is performed on the patent sequence P' (x) using Gompertz function. The Gompertz function is defined as:
where a, b, c are parameters of the curve, a is the asymptote, b controls the displacement of the curve on the x-axis, and c controls the growth rate.
Then, a sigmoidal curve g (x) of the maturation period is obtained.
FIG. 7 is an example of the fitting of a proprietary sequence by the technique "Mini computer", where the solid line is the original sequence and the dotted line is the fitting result.
Step 6, combining the two part curves
With the curves at the two stages, the maturity curves can be aligned in time and combined to generate. The last foam stage is known as [ st, ed]Its maximum value is located at
The maturity curve is then:
namely, the foam period curve is taken before a numerical point m, and the maximum value of the foam period curve and the maturation period curve after m indicates that the influence degree of the technology is dominated by the popular heat or market production at the time point.
The curve combination results of the two exemplary techniques are shown in fig. 8, in which the dotted line is the result of adaptive kernel density estimation of the document sequence, i.e. the bubble phase curve; the dotted line is the curve fitting result of the special sequence, namely a maturation period curve; the solid line is the technical maturity curve of the foam phase curve and the maturity curve.
To sum up, the method for automatically generating the technical maturity curve provided by the embodiment of the invention simulates the curve through nuclear density estimation and curve fitting by using historical data of papers and patents related to the technology, and obtains the technical maturity curve through synthesis, so that the maturity of the technology can be analyzed without manual operation and expert opinions, a certain reference value is provided for the state, enterprises and scientific researchers to judge the development condition of the technology, and the method is high in applicability, simple and easy to implement.
Next, an automatic generation device of a technical maturity curve proposed according to an embodiment of the present invention is described with reference to the drawings.
Fig. 9 is a schematic structural diagram of an apparatus for automatically generating a technical maturity curve according to an embodiment of the present invention.
As shown in fig. 9, the automatic technology maturity curve generation apparatus 10 includes: the system comprises an acquisition module 100, a foam period curve generation module 200, a maturity period curve generation module 300 and a maturity curve generation module 400.
The acquiring module 100 is configured to acquire a document sequence and a patent sequence; the foam period curve generation module 200 is configured to obtain a local bandwidth of the kernel density function according to the document sequence, estimate an adaptive kernel density of the document sequence according to the local bandwidth, and identify an inverse bell-shaped feature to generate a foam period curve; the maturity curve generating module 300 is configured to pre-process the patent sequence and perform curve fitting on the patent sequence to generate a maturity curve; the maturity curve generation module 400 is configured to generate a final maturity curve according to the foam period curve and the maturity period curve. The device 10 of the embodiment of the invention can analyze the maturity of the technology without manual operation and expert opinions, has a certain reference value for judging the development condition of the technology by countries, enterprises and scientific researchers, and has strong applicability and simple and easy realization.
Further, in one embodiment of the present invention, the definition formula of the local bandwidth is:
wherein, the dataPoint xiL (x) is a literature sequence, x ∈ { x ∈ }1，x2，...，xnS is a smoothing factor.
Further, in one embodiment of the present invention, the adaptive kernel density estimation formula is:
where K (x) is a kernel function.
Further, in an embodiment of the present invention, the maturity curve generating module 300 is further configured to perform an operation of accumulating the patent sequences to obtain an accumulated patent sequence, where the operation formula is:
wherein, p (x) is a patent sequence;
and carrying out data standardization on the accumulated patent sequences, wherein the standardization formula is as follows:
curve fitting the normalized patent sequence using Gompertz function, wherein the fitting formula is:
where a, b, c are parameters of the curve, a is the asymptote, b controls the displacement of the curve on the x-axis, and c controls the growth rate.
Further, in one embodiment of the present invention, the final maturity curve is:
wherein the content of the first and second substances,
m is a numerical point.
It should be noted that the foregoing explanation of the embodiment of the method for automatically generating a technology maturity curve is also applicable to the apparatus for automatically generating a technology maturity curve of this embodiment, and is not repeated here.
According to the automatic generation device of the technical maturity curve provided by the embodiment of the invention, the historical data of papers and patents related to the technology are used, the curve is simulated through nuclear density estimation and curve fitting respectively, and the technical maturity curve is obtained through synthesis, so that the maturity of the technology can be analyzed without manual operation and expert opinions, a certain reference value is provided for the state, enterprises and scientific researchers to judge the development condition of the technology, and the automatic generation device is high in applicability and simple and easy to implement.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (6)
1. A method for automatically generating a technical maturity curve is characterized by comprising the following steps:
acquiring a literature sequence and a patent sequence;
obtaining a local bandwidth of a kernel density function according to the document sequence, estimating an adaptive kernel density of the document sequence according to the local bandwidth, and identifying an inverted bell-shaped feature to generate a froth phase curve, wherein an estimation formula of the adaptive kernel density is as follows:
wherein K (x) is a kernel function;
preprocessing the patent sequence, and performing curve fitting on the patent sequence to generate a maturity curve; and
generating a final maturity curve according to the foam period curve and the maturity curve; wherein, the definition formula of the local bandwidth is as follows:
wherein the data point xiL (x) is a literature sequence, x ∈ { x ∈ }1，x2，...，xnS is a smoothing factor.
2. The method of claim 1, wherein the preprocessing the patent sequence numbers and curve fitting the patent sequences comprises:
and accumulating the patent sequence to obtain an accumulated patent sequence, wherein the operation formula is as follows:
wherein, p (x) is a patent sequence;
performing data normalization on the accumulated patent sequence, wherein the normalization formula is as follows:
curve fitting the normalized patent sequence using Gompertz function, wherein the fitting formula is:
where a, b, c are parameters of the curve, a is the asymptote, b controls the displacement of the curve on the x-axis, and c controls the growth rate.
3. The method of claim 2, wherein the final maturity curve is:
wherein the content of the first and second substances,
m is a numerical point.
4. An automatic generation device of a technical maturity curve is characterized by comprising:
the acquisition module is used for acquiring a document sequence and a patent sequence;
the foam period curve generation module is used for acquiring the local bandwidth of a kernel density function according to the document sequence, estimating the adaptive kernel density of the document sequence according to the local bandwidth, and identifying an inverted bell-shaped characteristic to generate a foam period curve, wherein an estimation formula of the adaptive kernel density is as follows:
wherein K (x) is a kernel function;
a maturity curve generating module, configured to pre-process the patent sequence and perform curve fitting on the patent sequence to generate a maturity curve; and
the maturity curve generation module is used for generating a final maturity curve according to the foam period curve and the maturity curve;
wherein, the definition formula of the local bandwidth is as follows:
wherein the data point xiL (x) is a literature sequence, x ∈ { x ∈ }1，x2，...，xnS is a smoothing factor.
5. The apparatus of claim 4, wherein the maturity curve generation module is further configured to perform an operation of accumulating the patent sequences to obtain an accumulated patent sequence, wherein the operation formula is:
wherein, p (x) is a patent sequence;
performing data normalization on the accumulated patent sequence, wherein the normalization formula is as follows:
curve fitting the normalized patent sequence using Gompertz function, wherein the fitting formula is:
where a, b, c are parameters of the curve, a is the asymptote, b controls the displacement of the curve on the x-axis, and c controls the growth rate.
6. The apparatus of claim 5, wherein the final maturity curve is:
wherein the content of the first and second substances,
m is a numerical point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911031409.9A CN110956675B (en) | 2019-10-28 | 2019-10-28 | Method and device for automatically generating technology maturity curve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911031409.9A CN110956675B (en) | 2019-10-28 | 2019-10-28 | Method and device for automatically generating technology maturity curve |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110956675A CN110956675A (en) | 2020-04-03 |
| CN110956675B true CN110956675B (en) | 2021-12-21 |
Family
ID=69975778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911031409.9A Active CN110956675B (en) | 2019-10-28 | 2019-10-28 | Method and device for automatically generating technology maturity curve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110956675B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768103A (en) * | 2020-06-30 | 2020-10-13 | 江苏赫玛信息科技有限公司 | Intelligent manufacturing capability maturity assessment method based on digital modeling |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6810332B2 (en) * | 2003-01-31 | 2004-10-26 | Chevron U.S.A. Inc. | Method for computing complexity, confidence and technical maturity indices for reservoir evaluations |
| CN102426670A (en) * | 2011-11-17 | 2012-04-25 | 重庆市科学技术研究院 | Technical value evaluation method and evaluation model of technical economic life |
| KR20140053444A (en) * | 2012-10-25 | 2014-05-08 | 한국과학기술정보연구원 | Apparatus for forecasting market volume, method of forecasting market volume, and storage medium for storing a program forecasting market volume |
| CN103208038B (en) * | 2013-05-03 | 2016-01-20 | 武汉大学 | A kind of patent introduces the computing method of predicted value |
| CN105184078A (en) * | 2015-09-06 | 2015-12-23 | 华南理工大学 | Technology maturity evaluation method based on patent relative-quantity analysis |
| CN106022616A (en) * | 2016-05-23 | 2016-10-12 | 北京航空航天大学 | Evaluation method for technical maturity of DC-DC power supply module |
| CN106384163A (en) * | 2016-09-08 | 2017-02-08 | 国网能源研究院 | New energy power generation technology maturity prediction method and device |
| CN110189016A (en) * | 2019-05-27 | 2019-08-30 | 中译语通科技股份有限公司 | Technology life cycle appraisal procedure and device |
-
2019
- 2019-10-28 CN CN201911031409.9A patent/CN110956675B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110956675A (en) | 2020-04-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109587713B (en) | Network index prediction method and device based on ARIMA model and storage medium | |
| CN104182474A (en) | Method for recognizing pre-churn users | |
| KR20160019897A (en) | Fast grouping of time series | |
| CN103778474A (en) | Resource load capacity prediction method, analysis prediction system and service operation monitoring system | |
| CN109743356B (en) | Industrial internet data acquisition method and device, readable storage medium and terminal | |
| CN112116198A (en) | Data-driven process industrial state perception network key node screening method | |
| CN115497272A (en) | Construction period intelligent early warning system and method based on digital construction | |
| CN112506752A (en) | Operation and maintenance method, device and equipment based on time series data trend prediction | |
| CN112631881A (en) | Operation and maintenance method, device and equipment based on time series data trend prediction | |
| CN112529053A (en) | Short-term prediction method and system for time sequence data in server | |
| CN110956675B (en) | Method and device for automatically generating technology maturity curve | |
| CN116384843B (en) | Energy efficiency evaluation model training method and monitoring method for digital energy nitrogen station | |
| CN111080037A (en) | Short-term power load prediction method and device based on deep neural network | |
| CN108427742B (en) | Power distribution network reliability data restoration method and system based on low-rank matrix | |
| CN112667394B (en) | Computer resource utilization rate optimization method | |
| CN113962741A (en) | Coal sales data prediction method, equipment and medium | |
| CN111985526B (en) | Similar scene clustering-based trailing interval management strategy generation method and system | |
| CN114282657A (en) | Market data long-term prediction model training method, device, equipment and storage medium | |
| CN115169089A (en) | Wind power probability prediction method and device based on kernel density estimation and copula | |
| CN114842178A (en) | Online visual interaction system and method based on electronic product | |
| CN113238908A (en) | Server performance test data analysis method and related device | |
| CN111340648A (en) | Energy management and control method and system based on Internet of things | |
| CN110222842A (en) | A kind of network model training method, device and storage medium | |
| CN112308446A (en) | Comprehensive judgment method, system and medium for technology maturity based on multivariate data | |
| CN110110906A (en) | A kind of survival risk modeling method based on Efron near-optimal |
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 |