CN113420946B - News media evaluation method - Google Patents
News media evaluation method Download PDFInfo
- Publication number
- CN113420946B CN113420946B CN202110076250.3A CN202110076250A CN113420946B CN 113420946 B CN113420946 B CN 113420946B CN 202110076250 A CN202110076250 A CN 202110076250A CN 113420946 B CN113420946 B CN 113420946B
- Authority
- CN
- China
- Prior art keywords
- index
- news
- evaluation
- indexes
- media
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 claims abstract description 53
- 238000013139 quantization Methods 0.000 claims abstract description 40
- 238000013179 statistical model Methods 0.000 claims abstract description 39
- 230000003993 interaction Effects 0.000 claims abstract description 31
- 230000007246 mechanism Effects 0.000 claims abstract description 28
- 238000013178 mathematical model Methods 0.000 claims abstract description 21
- 238000011002 quantification Methods 0.000 claims abstract description 12
- 238000007619 statistical method Methods 0.000 claims abstract description 12
- 238000011835 investigation Methods 0.000 claims description 19
- 238000003058 natural language processing Methods 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 5
- 230000009897 systematic effect Effects 0.000 abstract description 4
- 238000012216 screening Methods 0.000 abstract description 3
- 230000000875 corresponding effect Effects 0.000 description 21
- 238000004364 calculation method Methods 0.000 description 11
- 238000009954 braiding Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 4
- 238000005452 bending Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 241000272194 Ciconiiformes Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000003442 weekly effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
Abstract
Compared with the prior art, the method firstly acquires news stories and a plurality of interaction parameters from the electronic media, and processes the acquired data by constructing a plurality of professional news story mathematical models to obtain index scales of single news stories; then, carrying out statistical analysis and quantification treatment on the index table of the single news report by constructing an evaluation statistical model of the editing staff to obtain the index table of the editing staff; and finally, carrying out statistical analysis and quantization treatment on the index table of the editing personnel by constructing a media mechanism evaluation statistical model to obtain the index table of the media mechanism, thereby realizing the evaluation of the media mechanism. The method realizes the multidimensional, complete and systematic evaluation of the media mechanism, and the accuracy of the evaluation result is better than that of the existing method. The method can be used for performing performance assessment on mainstream news media, and can be used for accurately screening and selecting media practitioners.
Description
Technical Field
The invention relates to the technical field of media information processing, in particular to a method for evaluating news media.
Background
In the paper media age, the interaction attribute of newspaper release is lost, so that the works of news media practitioners cannot be truly and effectively read, counted and comment fed back, and the performance assessment of the news media practitioners is mainly carried out by the news reader to evaluate the manuscript text quality. With the global media becoming full-scale after 2010, relatively real and effective reading statistics and comment feedback can be obtained through novel electronic media such as websites and mobile clients, at present, few evaluation methods for professional states of news media practitioners remain in a stage of evaluating manuscript text quality by news reader, so that evaluation efficiency is low, and evaluation results are relatively subjective, so that research and development of an evaluation method and system for news media practitioners with systematicness, integrity, comprehensiveness and accuracy is a problem to be solved at present.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides a method for evaluating news media, which can realize multidimensional, complete and systematic evaluation of media institutions.
The invention provides a method for evaluating news media, which comprises the following steps:
acquiring a plurality of news reports published on an electronic medium in unit time and a plurality of interaction parameters corresponding to each news report;
constructing a plurality of news report evaluation mathematical models, wherein each news report evaluation mathematical model corresponds to one evaluation index, and acquiring a first index table corresponding to each news report from each news report and the corresponding interaction parameters according to the plurality of news report evaluation mathematical models;
constructing a collecting and editing personnel evaluation statistical model, and carrying out index quantification and statistics on all first index scales corresponding to the news reports of the collecting and editing personnel through the collecting and editing personnel evaluation statistical model to obtain second index scales corresponding to the collecting and editing personnel;
constructing a media mechanism evaluation statistical model, carrying out index quantization and statistics on a second index amount table corresponding to each editing staff through the media mechanism evaluation statistical model to obtain a third index amount table of the media mechanism, and realizing the evaluation of the media mechanism through the third index amount table.
According to the embodiment of the invention, at least the following technical effects are achieved:
compared with the existing quality evaluation scheme for manuscript texts by news readers, the method comprises the steps of firstly acquiring news stories and a plurality of interaction parameters corresponding to the news stories from an electronic medium in unit time as basic data for evaluating the single news stories, and processing the basic data of all selected single news stories by constructing a plurality of professional news story mathematical models to obtain a plurality of index quantitative indexes to form an index scale of the single news stories; then, carrying out statistical analysis and quantization treatment on index scales of all single news reports of the editing personnel by constructing an evaluation statistical model of the editing personnel to obtain index quantization indexes of the editing personnel, and forming the index scales of the editing personnel; and finally, carrying out statistical analysis and quantization treatment on index tables of all editing staff by constructing a media mechanism evaluation statistical model to obtain index quantization indexes of the media mechanism, and forming the index tables of the media mechanism, thereby realizing evaluation of the media mechanism. The method realizes the multidimensional, complete and systematic evaluation of the media mechanism, and the accuracy of the evaluation result is better and more objective than that of the existing method. The method can be used for performing performance assessment on mainstream news media, can be widely used for units and individuals with transmission requirements and reporting requirements, and can be used for accurately screening and selecting media practitioners.
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 invention will become apparent and may be better understood from the following description of embodiments taken in conjunction with the accompanying drawings in which:
fig. 1 is a flow chart of a method for evaluating news media according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a trending topic provided by an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a news media evaluation device according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
Referring to fig. 1, an embodiment of the present invention provides a method for evaluating news media, including the steps of:
s100, acquiring a plurality of news stories published on an electronic medium in unit time and a plurality of interaction parameters corresponding to each news story.
S200, constructing a plurality of news story evaluation mathematical models, wherein each news story evaluation mathematical model corresponds to one evaluation index, and acquiring a first index table corresponding to each news story from each news story and a plurality of corresponding interaction parameters according to the plurality of news story evaluation mathematical models.
S300, constructing a collecting and editing personnel evaluation statistical model, and carrying out index quantification and statistics on first index scales corresponding to all news reports of the collecting and editing personnel through the collecting and editing personnel evaluation statistical model to obtain second index scales corresponding to the collecting and editing personnel.
S400, constructing a media mechanism evaluation statistical model, carrying out index quantification and statistics on a second index amount table corresponding to each editing staff through the media mechanism evaluation statistical model to obtain a third index amount table of the media mechanism, and realizing evaluation of the media mechanism through the third index amount table.
Compared with the prior quality evaluation scheme for manuscript text by news readers, the method firstly obtains news reports and a plurality of interaction parameters corresponding to the news reports from electronic media in unit time as basic data for evaluating single news reports, wherein the electronic media can be various large websites (such as WeChat public numbers, microblogs, today's headlines, penguin numbers, hundreds of families, internet easily numbers, search fox numbers, large fish numbers, large wind numbers, new wave networks and other self-media information platforms and various main stream media official websites) or mobile clients, and the interaction parameters comprise but are not limited to reading numbers, praise numbers, comment numbers and interaction rates on the electronic media, and the basic data of all selected single news reports are processed by constructing a plurality of professional news report mathematical models to obtain a plurality of index quantization indexes to form an index scale of the single news report; then, carrying out statistical analysis and quantization treatment on index scales of all single news reports of the editing personnel by constructing an evaluation statistical model of the editing personnel to obtain index quantization indexes of the editing personnel, and forming the index scales of the editing personnel; and finally, carrying out statistical analysis and quantization treatment on index tables of all editing staff by constructing a media mechanism evaluation statistical model to obtain index quantization indexes of the media mechanism, and forming the index tables of the media mechanism, thereby realizing evaluation of the media mechanism. The method realizes the multidimensional (multiple evaluation indexes), complete and systematic evaluation of the media mechanism, and the accuracy of the evaluation result is better and more objective than that of the existing method. The method can be used for performing performance assessment on mainstream news media, can be widely used for units and individuals with transmission requirements and reporting requirements, and can be used for accurately screening and selecting media practitioners.
As an alternative embodiment, the evaluation indexes in the first index table include an influence index, a depth index, a readability index, a story index, a survey or value trend index, and a serious or liveness trend index; the evaluation indexes in the second index scale comprise an influence index, a depth index, a readability index, a story index, a investigation or value trend index, a serious or liveness trend index, a topic concentration index and a line mouth concentration index; the evaluation indexes in the third index scale comprise an influence index, a depth index, a readability index, a story index, a investigation or value trend index, a serious or liveness trend index, a topic concentration index, a line mouth concentration index, a gatherer stability index and a hot spot tracking index. As shown in table 1 below:
index name | News report | Collecting and braiding personnel | Media mechanism | |
1 | Influence of force | Function calculation → | Function derivation → | Statistical model |
2 | Depth of depth | Statistics of NLP → | Statistical model → | Statistical model |
3 | Readability of | NLP→ | Statistical model → | Statistical model |
4 | Story of telling | NLP→ | Statistical model → | Statistical model |
5 | Investigation or value inclinationTo the direction of | NLP→ | Statistical model → | Statistical model |
6 | Serious or active tendency | NLP→ | Statistical model → | Statistical model |
7 | Topic concentration | Statistics of NLP → | Statistical model | |
8 | Line mouth concentration degree | Statistics of NLP → | Statistical model | |
9 | Stability of gatherer | Statistical model | ||
10 | Hot spot tracking degree | Statistical model |
TABLE 1
NLP (Natural Language Processing ). Compared with the prior technical scheme that the quality of the news report is evaluated only by news readers, the method of the embodiment evaluates all single news reports in unit time by acquiring quantitative indexes of multiple dimensions, wherein the evaluation dimension comprises index dimensions (i.e. influence indexes) for representing the level and the capability of the single news report and index dimensions (i.e. indexes such as darkness, readability, storyliness, investigation or value tendency, serious or liveness tendency and the like) for representing the characteristic index of the single news report, so that the evaluation of the single news report is more accurate and objective and has more practical application value. According to the method, after all the single news reports are evaluated correspondingly through a professional mathematical model, a single news report index table is obtained, then the single news report index table is used as basic data, the single news report index table is collected under the name of a corresponding editing personnel to carry out statistical calculation, topic concentration index and line mouth concentration index are increased, comprehensive modeling calculation is carried out, an evaluation index table of the editing personnel is obtained, and the increased topic concentration index and line mouth concentration index can enable evaluation of the editing personnel to be more accurate and objective. And finally, carrying out index quantification on the evaluation index table of the editing personnel through a professional evaluation model to obtain the evaluation table of the media institution, and simultaneously, increasing two indexes of stability and hot spot tracking degree of the editing personnel in the process of generating the evaluation table of the media institution by the model so as to increase the evaluation accuracy.
Wherein, the stability index and the hot spot tracking index of the gatherer comprise:
stability index of the gatherer: according to the personnel retention number in the time-sharing interval, referring to the total number of personnel and the change value, and carrying out weighted calculation on personnel stability;
hot spot tracking index: the hot spot tracking degree is an index obtained by carrying out statistical calculation on a reported tracking index of a media in a hot spot news event in a certain time interval.
The quantization process of each index is described below:
based on the above embodiment, constructing a specific news story evaluation mathematical model for quantifying an impact index includes the steps of:
s1011, carrying out standardization processing on a plurality of interaction parameters to obtain standard values of the interaction parameters.
The interaction parameters include reading numbers, praise numbers, comment numbers and interaction rates for a single news story on the electronic media. The aim of the standardized processing of the interaction parameters is to remove the inconsistency of the interaction parameters, so that the interaction parameters can be compared and calculated.
S1012, obtaining information entropy of the interaction parameters, and calculating weights of the interaction parameters according to the information entropy of the interaction parameters.
S1013, calculating the quantization index of the influence index according to the standard values of the interaction parameters and the weights of the interaction parameters.
Based on the present embodiment, a periodicity factor may also be added in the process of the quantitative calculation of the impact index, and it should be noted that the periodicity factor herein is different from the unit time described in step S100 of the above embodiment. The added periodicity factor has the advantages that the influence condition of the single news report in the corresponding period can be reflected, and the method can also be used for cross-period comparison of the influence condition of the single news report.
To facilitate understanding of the above steps S1011 to S1013 by those skilled in the art, a set of examples are listed below, and it should be noted that a periodicity factor is added in this example:
(1) Normalizing k index parameters;
suppose k index parameters are selected: x is X 1 ,X 2 ,...,X k Wherein X is i ={X 1 ,X 2 ,...,X k }. Let n=1, 7, 30, 356, respectivelyCorresponding to daily, weekly, monthly, yearly, etc. periods.
Assume that the cardinality of the k index parameters is:the base comparison is established, the base is like an index scale, the relative positions of k index parameters can be intuitively displayed, the k index parameters are subjected to standardization processing, and the obtained standard values are as follows: y is Y 1 ,Y 2 ,...,Y k Wherein Y is i ={Y 1 ,Y 2 ,...,Y k }。
Then:
setting up the upper limit of the standardized value to be 1000, and obtaining the following steps:
Y i =(Y i >1000)?1000:Y i 。
(2) Obtaining the information entropy of k index parameters;
according to the definition of the information entropy in the information theory, the information entropy is as follows:
wherein,if p is i =0, then define: />
(3) The weight of k index parameters is obtained;
according to the calculation formula of the information entropy, the information entropy of each index is calculated as follows: e (E) 1 ,E 2 ,...,E k Wherein E is i ={E 1 ,E 2 ,…,E k }。
Calculating the weight of each index through information entropy:
(4) Obtaining a quantization index of the influence dimension;
where NI represents the quantization index of the influence dimension.
Based on the above embodiment, constructing a specific news story evaluation mathematical model to quantify the darkness index includes the steps of:
s1021, calculating a logic complexity index and a space index from the news report.
(1) A logical complexity index for news stories;
the method comprises the steps of performing word segmentation, part of speech judgment and other processing on texts of news reports by using NLP, extracting a message source main body by context analysis, counting total main body quantity T and main body total frequency P, analyzing and counting average message source main body quantity AT of all news reports in a database, taking the average message source main body quantity AT as a judgment base of the total main body quantity T of the news, squaring the ratio of the main body quantity, increasing the influence of news with large main body quantity on indexes, and reducing the influence of news with small main body quantity. The highest subject total frequency Pm is established. Meanwhile, the distribution situation of each main body in the text is counted in a segmented mode, the main body distribution situation of the news report is represented by a main body distribution index D, if every 500 words are set as one segment, the news is divided into n segments, and the main body quantity of each segment is Tn:
where L represents the logical complexity index of the news story.
(2) A spread index;
the word number W of the news report is used for judging the depth of the news reportOne of the core elements of the shallowness index. After the database news report data is processed by eliminating abnormal values, the data is sequenced from high to low according to word numbers, and the average word number of the first 1000 news reports like the top of a pyramid is taken as the highest news word number constant W m :
Wherein S represents the spread index.
(3) A quantization index of the depth index;
obtaining a logic complexity index weight W by an entropy weight method mentioned by quantification of the influence indexes l Sum spread index weight W s Finally, the quantization index H of the depth index is obtained by the following formula:
H=L*W l +S*W s
based on the above embodiments, constructing a specific news story evaluation mathematical model quantifies the readability index comprising the steps of:
s1031, performing professional term word stock collision on the news report, and performing long short sentence division by using a natural language processing method.
S1032, calculating the quantization index of the readability index based on the result of the collision library and the long sentence division.
For example, news stories are more in terms of terms, exceeding a set threshold (freely settable), with a low number of readability indicators, otherwise the article is of the popular and understandable type. The long short sentence division refers to judging the length of the paragraph and the sentence structure of the article, for example, the length of the paragraph and the length of the long sentence represent poor readability, and the readability index number is a low value.
As an alternative embodiment, constructing a specific news story evaluation mathematical model quantifies a story indicator comprising the steps of:
s1041, acquiring space-time conversion times, character occurrence frequency, contradiction conflict frequency, zigzag frequency and verb distribution frequency from news reports through a natural language processing method and a statistical method.
Wherein the variables involved in the story index include:
number of space-time conversions: the number of times and places appearing in the news report and the distribution value, the number of times and the distribution uniformity of the text distance of the times and places; the time-space conversion times are positively correlated with the story index;
personage (principal angle) frequency of occurrence: the number of occurrences of different person names, the number of occurrences of each person and the number of occurrences of distribution transformation form positive correlations according to different weights, and parameter weighting calculation can be performed by adopting a function calculation method;
contradictory conflict frequency: in news report, the occurrence frequency of the turning conjunctions and the occurrence frequency of the negatives are used for carrying out weighted calculation, and the conjunctions with different collision degrees are respectively assigned, for example, however, the disjunctive weights are nevertheless, etc., and the negated word weights are represented, for example, forbidden, impermissible, inaccurate, inadvisable, otherwise, etc. The weighting calculation may also employ a complex function.
Meandering frequency: the bending sense is measured by the different combination quantity of space time and principal angles, when a space time expression appears, the situation that the character closest to the space time expression appears, and the different quantity of the combination forms the space time bending combination of the character, and the bending frequency is positively related to the story index;
distribution frequency of verbs: the number of verbs appearing in news stories is in positive correlation with the difference distribution of verbs, and the two values are in positive correlation with the story index.
S1042, calculating quantization index of story index based on the obtained time-space conversion times, character occurrence frequency, contradiction conflict frequency, tortuous frequency and verb distribution frequency.
The quantization index of the story index can be obtained through a weighting operation.
Based on the above embodiments, constructing a specific news story evaluation mathematical model quantifies a survey or value trend indicator includes the steps of:
s1051, dividing news stories into investigation articles and value articles.
S1052, obtaining survey word stock and value word stock corresponding to the survey articles and the value articles respectively according to the word frequency statistical method.
S1053, based on the survey word stock and the value word stock, the survey degree score and the value degree score are carried out on the survey article and the value article through a natural language processing method.
S1054, calculating quantitative indexes of investigation or value trend indexes according to the investigation degree scores and the value degree scores.
One set of examples is listed below:
500 investigation articles and value articles are respectively selected, wherein the two articles are obtained by manual scoring. Through word segmentation, part-of-speech filtering and related processing, a survey class and value class word library is established.
Taking the value class word library as an example, table 2 below shows one value class word library:
word name | Part of speech | Weight summary | Weighted average |
Benefit of | n | [ { times: 2, weight: 9 }.] | 8.5 |
Profit margin | n | [ { times: 1, weight: 8 }.] | 7.9 |
… | … | … | … |
TABLE 2
Based on the value class word stock, determining a highest score threshold, taking the value as a judgment standard, and if the value is higher than the value, fully dividing the value into 100%, otherwise gradually decreasing the value index:
for example:
description of the form: word name: the number of times: weighting of
Value word stock: the yield is 1:8.5; profit 1:7.9
Let the highest sub-threshold be 20:
finally, the value index and the investigation index (investigation index is not shown in the embodiment) are summarized to obtain the quantitative index of investigation or value trend index.
Based on the above embodiment, constructing a specific news report evaluation mathematical model to quantify serious or liveness trend indicators includes the following steps:
s1061, dividing news stories into serious articles and active articles.
S1062, acquiring a serious class word stock and an active class word stock corresponding to all serious class articles and all active class articles respectively according to a word frequency statistical method.
S1063, scoring the seriousness and the liveness corresponding to the seriousness articles and the liveness articles by a natural language processing method based on the seriousness word stock and the liveness word stock.
S1064, calculating a quantization index of the serious or liveness trend index according to the serious degree score and the liveness degree score.
One set of examples is listed below:
similar to the quantization process of the similar investigation or the value trend index, 500 articles of serious class (immediate administration class) and active class (entertainment class) are respectively selected, and through word segmentation, part-of-speech filtering and related processing, a word library of the serious class and the active class is established.
Taking the serious class word library as an example, table 3 below shows one serious class word library:
word name | Part of speech | Weight summary | Weighted average |
Politics | n | [ { times: 2, weight: 9 }.] | 8.5 |
... | ... | ... | ... |
... | ... | ... | ... |
TABLE 3 Table 3
And (3) respectively carrying out serious degree scoring and liveness scoring on the single news report based on the serious class word stock and the liveness class word stock, and calculating the quantization index of serious or liveness tendency dimension according to the bipolar scoring. The quantization index of the serious or liveness tendency dimension is based on the serious degree score and the liveness degree score, a two-pole model is built according to the two scores, serious class is left pole=0, liveness class is right pole=1, the index is more prone to the serious class when being close to 0, and otherwise, the index is more prone to the liveness class.
For example:
description of the form: word name: the number of times: weighting of
The scoring situation of the news report is as follows:
serious score = 1 x 9+1 x 7 = 16
Liveness score = 1*8 = 8
The index is near 0 and the news report belongs to serious article.
Finally, the quantitative index of serious or active tendency index of the news report is obtained in a summarizing way.
Based on the above embodiments, as in table 1 above, the quantification of the panel evaluation statistical model dialog topic concentration index and line mouth concentration index includes the following steps:
s2071, quantifying the topic concentration index.
Acquiring a plurality of topics from news reports of the editing staff to cluster and counting the times of each topic;
selecting a plurality of topics with top frequency ranking, and taking the proportion of the selected topics with top frequency ranking in all topics as a quantization index of the topic concentration index.
S2072, quantifying the line port concentration index.
Acquiring the corresponding duty ratio of each line port from the news report of the editing staff;
and calculating the quantization index of the line port concentration index according to the maximum line port duty ratio and the variance of the duty ratio of each line port.
S2073, performing weighting operation on the obtained topic concentration degree and line mouth concentration degree, and performing weighting operation by combining the obtained index quantization indexes to obtain an index table for the editing personnel.
One set of examples is listed below:
(1) Referring to fig. 2, the first 20 to 30 trending topics of a certain gatherer are extracted for clustering and the number of statistical topics are combined. The method of clustering used in this example is word distance clustering. The k-means clustering method may also be used for optimization here.
(2) Extracting the first three types of topics to calculate topic concentration:
for example:
quantization index of topic concentration of gatherer = (283+42+30)/566= 62.72%, where 283, 42 and 30 are the number of times of the first three categories of topics, respectively; 566 is the total number of topics for the news story.
(3) Counting the number of news stories of the gatherer on each line mouth and the total number of the news stories, wherein:
and obtaining the news report duty ratio of the editing staff under each line port.
(4) Taking the maximum line port duty ratio as a main index, and adding variance (indicating data stability) to comprehensively judge the line port concentration degree:
the following table 4 gives the data of the line mouth concentration quantization index for the braiding persons a to F:
time duty ratio | Military duty cycle | Entertainment duty cycle | Agricultural duty cycle | Line mouth concentration degree | |
Collecting and braiding personnel A | 25% | 25% | 25% | 25% | 25.00% |
Staff B | 50% | 20% | 20% | 10% | 50.23% |
Collecting and braiding person C | 70% | 10% | 10% | 10% | 70.67% |
Collecting and braiding person D | 70% | 20% | 10% | \ | 70.69% |
Collecting and braiding person E | 90% | 10% | \ | \ | 91.60% |
Collecting and braiding person F | 100% | \ | \ | \ | 100.00% |
TABLE 4 Table 4
(5) And carrying out weighted operation on the calculated quantization index of the topic concentration index, the quantization index of the line mouth concentration index and the calculated quantization indexes of each index to obtain an index table aiming at the editing staff.
Based on all the above embodiments, each model (including a plurality of news report evaluation mathematical models, a gatherer evaluation statistical model and a media agency evaluation statistical model) realizes function tuning through a deep learning method. The method can also introduce manual experts to perform posterior parameter adjustment at the same time, and aims to continuously correct functions so as to improve the fitting degree of the model, so that the model has a continuously improved function, and the method can realize self-adaptive adjustment according to actual conditions and ensure the accuracy of evaluation.
Referring to fig. 3, in one embodiment of the present invention, a news media evaluation device is provided, which may be any type of intelligent terminal, such as a mobile phone, a tablet computer, a personal computer, etc. Specifically, the news media evaluation device includes: one or more control processors and memory, here exemplified by one control processor. The control processor and the memory may be connected by a bus or otherwise, here by way of example.
The memory is used as a non-transitory computer readable storage medium, and can be used for storing non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions/modules corresponding to the evaluation device of the news media in the embodiment of the invention. The control processor executes the non-transitory software program, instructions and modules stored in the memory, thereby implementing the method for evaluating news media according to the above embodiment.
The memory may include a memory program area and a memory data area, wherein the memory program area may store an operating system, at least one application program required for a function; the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the control processor, the remote memory being connectable to the evaluation device of the news media via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The one or more modules are stored in the memory and when executed by the one or more control processors, perform the method of evaluating news media described in the above embodiments.
The present invention also provides a computer-readable storage medium storing computer-executable instructions for performing the method for evaluating news media according to the above embodiments by one or more control processors.
From the above description of embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented in software plus a general purpose hardware platform. Those skilled in the art will appreciate that all or part of the flow of the method of the above-described embodiments may be implemented by a computer program to instruct related hardware, and the program may be stored in a computer readable storage medium, and the program may include the flow of the embodiment of the method as described above when executed. The storage medium may be a magnetic disk, an optical disk, a Read Only Memory (ROM), a random access Memory (Random Access Memory, RAM), or the like.
In the description of the present specification, reference to the terms "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., means 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, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
Claims (7)
1. A method for evaluating news media, comprising the steps of:
acquiring a plurality of news reports published on an electronic medium in unit time and a plurality of interaction parameters corresponding to each news report;
constructing a plurality of news report evaluation mathematical models, wherein each news report evaluation mathematical model corresponds to one evaluation index, and acquiring a first index table corresponding to each news report from each news report and the corresponding interaction parameters according to the plurality of news report evaluation mathematical models;
constructing a collecting and editing personnel evaluation statistical model, and carrying out index quantification and statistics on all first index scales corresponding to the news reports of the collecting and editing personnel through the collecting and editing personnel evaluation statistical model to obtain second index scales corresponding to the collecting and editing personnel;
constructing a media mechanism evaluation statistical model, carrying out index quantization and statistics on a second index amount table corresponding to each editing staff through the media mechanism evaluation statistical model to obtain a third index amount table of the media mechanism, and realizing the evaluation of the media mechanism through the third index amount table;
the news report evaluation mathematical model, the editing personnel evaluation statistical model and the media mechanism evaluation statistical model are all used for realizing function parameter adjustment by a deep learning method or introducing artificial experts simultaneously for posterior parameter adjustment;
the evaluation indexes in the first index table comprise influence indexes, depth indexes, readability indexes, story indexes, investigation or value trend indexes and serious or liveness trend indexes; the evaluation indexes in the second index table comprise influence indexes, depth indexes, readability indexes, story indexes, investigation or value trend indexes, serious or liveness trend indexes, topic concentration indexes and line mouth concentration indexes; the evaluation indexes in the third index table comprise an influence index, a depth index, a readability index, a story index, a investigation or value trend index, a serious or liveness trend index, a topic concentration index, a line mouth concentration index, a gatherer stability index and a hot tracking index;
wherein the quantifying of the impact indicator comprises the steps of:
carrying out standardization processing on a plurality of interaction parameters in a corresponding period to obtain standard values of the interaction parameters;
acquiring information entropy of the interaction parameters, and calculating weights of the interaction parameters according to the information entropy of the interaction parameters;
and calculating the quantization index of the influence index according to the standard values of the interaction parameters and the weights of the interaction parameters.
2. The method of evaluating news media according to claim 1, wherein the quantifying of the darkness index comprises the steps of:
calculating a logical complexity index and a spread index from the news stories;
and calculating a quantization index of the depth index according to the calculated logic complexity index and the calculated space index.
3. The method of evaluating news media according to claim 1, wherein the quantifying of the readability index comprises the steps of:
performing professional term word stock collision on news reports, and performing long and short sentence division by using a natural language processing method;
and calculating the quantization index of the readability index based on the result of the collision library and the long sentence division.
4. The method of evaluating news media according to claim 1, wherein the quantifying of the story index comprises the steps of:
acquiring space-time conversion times, character occurrence frequency, contradiction conflict frequency, zigzag frequency and verb distribution frequency from news reports through a natural language processing method and a statistical method;
and calculating a quantization index of the story index based on the acquired number of space-time conversions, the character occurrence frequency, the contradiction conflict frequency, the meandering frequency and the distribution frequency of the verbs.
5. The method of evaluating news media according to claim 1, wherein the quantifying of the survey or value trend indicators comprises the steps of:
dividing news stories into investigation class articles and value class articles;
respectively acquiring a survey class word stock and a value class word stock corresponding to the survey class articles and the value class articles according to a word frequency statistical method;
based on the investigation class word stock and the value class word stock, carrying out investigation degree scoring and value degree scoring on the investigation class articles and the value class articles by a natural language processing method;
and calculating a quantitative index of the survey or the value trend index according to the survey grade and the value grade.
6. The method of evaluating news media according to claim 1, wherein the quantification of the serious or liveness tendency index comprises the steps of:
dividing news reports into serious articles and active articles;
acquiring a serious class word stock and an active class word stock corresponding to all the serious class articles and all the active class articles respectively according to a word frequency statistical method;
based on the serious class word stock and the active class word stock, scoring the serious degree and the active degree corresponding to the serious class articles and the active class articles by a natural language processing method;
and calculating a quantization index of the serious or liveness tendency index according to the serious degree score and the liveness degree score.
7. The method for evaluating news media according to claim 1, wherein the quantification of the topic concentration index and the line mouth concentration index by the gatherer evaluation statistical model comprises the steps of:
quantification of topic concentration index:
acquiring a plurality of topics from news reports for clustering and counting the times of each topic; selecting a plurality of topics with top frequency ranking, and taking the proportion of the selected topics with top frequency ranking in all topics as a quantization index of a topic concentration index;
quantification of line mouth concentration index:
acquiring the corresponding duty ratio of each line port from the acquired news report;
and calculating the quantization index of the line port concentration index according to the maximum line port duty ratio and the variance of the duty ratio of each line port.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110076250.3A CN113420946B (en) | 2021-01-20 | 2021-01-20 | News media evaluation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110076250.3A CN113420946B (en) | 2021-01-20 | 2021-01-20 | News media evaluation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113420946A CN113420946A (en) | 2021-09-21 |
CN113420946B true CN113420946B (en) | 2024-02-09 |
Family
ID=77711649
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110076250.3A Active CN113420946B (en) | 2021-01-20 | 2021-01-20 | News media evaluation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113420946B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114036221A (en) * | 2021-09-24 | 2022-02-11 | 国务院国有资产监督管理委员会研究中心 | Thematic event analysis method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346425A (en) * | 2014-07-28 | 2015-02-11 | 中国科学院计算技术研究所 | Method and system of hierarchical internet public sentiment indication system |
CN106294334A (en) * | 2015-05-11 | 2017-01-04 | 国家计算机网络与信息安全管理中心 | The computational methods of a kind of microblogging public sentiment index system and device |
CN109359857A (en) * | 2018-10-12 | 2019-02-19 | 网智天元科技集团股份有限公司 | A kind of influence of media force estimation method, apparatus and electronic equipment |
CN109800349A (en) * | 2018-12-17 | 2019-05-24 | 北京邮电大学 | The data processing method and device of content quantization news value are issued based on user |
WO2020000847A1 (en) * | 2018-06-25 | 2020-01-02 | 中译语通科技股份有限公司 | News big data-based method and system for monitoring and analyzing risk perception index |
CN111552882A (en) * | 2020-05-09 | 2020-08-18 | 重庆邮电大学 | News influence calculation method and device, computer equipment and storage medium |
-
2021
- 2021-01-20 CN CN202110076250.3A patent/CN113420946B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346425A (en) * | 2014-07-28 | 2015-02-11 | 中国科学院计算技术研究所 | Method and system of hierarchical internet public sentiment indication system |
CN106294334A (en) * | 2015-05-11 | 2017-01-04 | 国家计算机网络与信息安全管理中心 | The computational methods of a kind of microblogging public sentiment index system and device |
WO2020000847A1 (en) * | 2018-06-25 | 2020-01-02 | 中译语通科技股份有限公司 | News big data-based method and system for monitoring and analyzing risk perception index |
CN109359857A (en) * | 2018-10-12 | 2019-02-19 | 网智天元科技集团股份有限公司 | A kind of influence of media force estimation method, apparatus and electronic equipment |
CN109800349A (en) * | 2018-12-17 | 2019-05-24 | 北京邮电大学 | The data processing method and device of content quantization news value are issued based on user |
CN111552882A (en) * | 2020-05-09 | 2020-08-18 | 重庆邮电大学 | News influence calculation method and device, computer equipment and storage medium |
Non-Patent Citations (2)
Title |
---|
基于AHP 法的个人博客影响力评价方法研究;朱丽等;《数学的实践与认识》;20080831;第38卷(第15期);第109-117页 * |
新媒体环境下新闻传播效果评估的指标和权重;刘建明等;《新闻与传播评论》;20180930;第71卷(第4期);第41-57页 * |
Also Published As
Publication number | Publication date |
---|---|
CN113420946A (en) | 2021-09-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10878004B2 (en) | Keyword extraction method, apparatus and server | |
CN110297988B (en) | Hot topic detection method based on weighted LDA and improved Single-Pass clustering algorithm | |
Gilbert et al. | Widespread worry and the stock market | |
CN110619506B (en) | Post image generation method, post image generation device and electronic equipment | |
WO2017067153A1 (en) | Credit risk assessment method and device based on text analysis, and storage medium | |
CN106933806A (en) | The determination method and apparatus of medical synonym | |
CN108133418A (en) | Real-time credit risk management system | |
CN106919689A (en) | Professional domain knowledge mapping dynamic fixing method based on definitions blocks of knowledge | |
WO2018184518A1 (en) | Microblog data processing method and device, computer device and storage medium | |
TW201915777A (en) | Financial analysis system and method for unstructured text data | |
CN106776672A (en) | Technology development grain figure determines method | |
CN108363699A (en) | A kind of netizen's school work mood analysis method based on Baidu's mhkc | |
CN113420946B (en) | News media evaluation method | |
CN109086794A (en) | A kind of driving behavior mode knowledge method based on T-LDA topic model | |
CN116611546A (en) | Knowledge-graph-based landslide prediction method and system for target research area | |
CN110019763B (en) | Text filtering method, system, equipment and computer readable storage medium | |
CN109979592A (en) | Mental health method for early warning, user terminal, server and system | |
Hantke et al. | Trustability-based dynamic active learning for crowdsourced labelling of emotional audio data | |
CN111931031A (en) | Method for calculating policy information matching degree | |
CN115738285A (en) | Game quality evaluation feedback method and system | |
CN114077977B (en) | Building intelligent management method and system based on big data and readable storage medium | |
Zhu et al. | Research on data mining of college students’ physical health for physical education reform | |
CN111767404B (en) | Event mining method and device | |
CN113988878A (en) | Graph database technology-based anti-fraud method and system | |
CN105975792A (en) | Big-data-based dermatoglyph analysis and processing device and method |
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 |