0　Introduction

With the rapid development of science and technology, intellectual property gains increasing attentions. Large amount of patent information with detailed content and standardized format produced in this process follows the law of information obsolescence like other scientific literature and network information. Meanwhile, multi-level qualification examination of patents guarantees the novelty, practicability and creativity of patent information while their timeliness insures that patent literature can present technological innovation at the soonest and objectively reflects the innovation trend of technology as well as the strength of technological innovation of related enterprises. The measure for the life span of patent information becomes a new perspective to grasp the trend of technological innovation and evaluate the competitiveness of technology-based enterprises.

Wireless charging is a relatively new and vibrating industry, however, its related patents with classic feature can be traced back long time ago. Hence, ‘double-proportion’ obsolescence method is adopted to measure the life span of patents in the field of wireless charging, and its law of development combined with examples is further explored with in-depth analysis.

1　Measurement of patent life span

1.1　Methodsofthemeasurementofpatentlifespan

The patent life span, studying patent life cycle from the time dimension, refers to the length of time lasting from the time when a patent being authorized to the time when the patent value declines to infinite low. In the research of the measurement of patent life span, the core issue is how to determine the obsolescence point of a patent. To sum up, there are mainly three methods to determine the obsolescence point in recent years:

(3)宣传力度不够。整体上长沙有一张闪亮的城市旅游名片，但是望城区、浏阳市、宁乡县和长沙县的单体旅游景点还只是在初步的开发当中，当前旅游产品总量上偏少，且层次不够丰富，缺乏突出的旅游龙头项目。比如长沙县有浔龙河特色小镇与乡村旅游深度融合的现代休闲模式，有白沙水果园、春华荷花节等休闲农业旅游模式等，但始终没有主打品牌，没有龙头项目，有景点但没有亮点，呈现散、小、乱、低端的状态。

Firstly, the life span of information can be determined by half-life of an information set and the Price index. Academics including Harter[1], Bugeja[2], and Koehler[3] taking websites or web pages as objects, refined ‘link decay’ as the measurement standard of information obsolescence and then calculated the half-life of the ‘link obsolescence’ in an information set, that is, the time during which half of links in the set has obsolesced. Due to the particularity of patent information, half-life of the sets and the Price index are inappropriate to the determination of individual obsolescence of information.

Secondly, the information can be deemed to be obsolesced when the utilization frequency reaches a certain proportion P. This method is an extension to the research method of individual half-life , and it is not feasible since every single information needs to be tracked until it is unattended.

3.坚持全球视野。我国的乡村振兴是全球乡村治理体系不可或缺的一个重要组成部分，乡村振兴规划要坚持国际视野，努力在全球乡村治理体系中贡献中国方案和中国智慧。[2]

Thirdly, the information without any users’ attention within a consecutive period T can be considered obsolesced. Wang,et al.[4] defined the life cycle of network information as a time span from the generation of network information to the stage when no users used this information, using page views to depict the process. The research results indicate that around 77% of web pages achieve their maximum in terms of users’ interests on the first day after their birth, and for the rest users, their interests’ peak within one week. In addition, the active period of more than 80% of the web pages with timeliness has lasted no longer than two weeks. However, it is unrealistic to judge whether the information is completely unattended, as the determination of the T value has become a difficulty due to the lack of the support of scientific methods, and is mostly set by experience.

1.2　The introduction of ‘double-proportion’ obsolescence method of life span

As economic index is hard to get, academics mostly measure the value of patent by its citation[5]. Therefore, ‘double-proportion’ obsolescence adopted in this paper also employs the patents’ citation as data basis to determine the obsolescence point of a patent[6]. The method combines thoughts of two methods, respectively cumulative citation percentage and continuous silent time, to determine the single patent obsolescence time with considering the proportion of obsolesced sample patents in the total sample patents. The method plays an important role when determining the obsolescence point of network information on the NetEase news, the Strong China Forum, and so on. It can also be used to determine the measure of obsolescence in other types of information[7] according to the difference of information life span by adjusting the P value.

与传统电力营销方式相比，基于费控策略的电力营销能够有效缩短电费回收期，传统营销模式下，电费回收包括了人工抄表、后台系统核算以及人工收费等，人工抄表受外界的因素影响较大，在时间上存在很大的差异，本文以石拐供电分局传统抄核收情况为例分析，一般供电企业抄表时间为每月10-14日，核算 15-17日，收费时间 18－28日，需要时间较长，而采用费控模式进行电费收缴，时间会大大缩短，因为费控营销系统下的抄表、结算以及收费均能在系统中自行完成，仍以包头供电局为例，费控策略下的收费时间为每月10日，核算11-12日，收费一般2-3天就能完成，传统模式下和费控模式下的电费回收对比情况如表1所示。

In the fields of library and information, many scholars have different understandings and applications on time delay. Scholars like Wang Yuandi pointed out that the ‘citation delay’ of a paper refers to the delay between its publication date of this paper and the publication date of the article citing this paper[9]. But Han Zhiwei pointed out that readers benefited from reading the papers, the time lapse from then till the project has papers being published and new papers being cited, can be referred to as ‘citation delay’[10]. It can be seen that according to the different contents of the study, different scholars have various determinations of the time difference called ‘citation delay’, but all reflect the timeliness of the papers. In this paper, the patent citation delay is defined as: the time difference between the patent authorization and its first time being cited.

(2) Calculate the citation frequency Nij of each patent in each period during the calculating time span, where i(1≤i≤N) represents the patent, j represents the time period, and j′ represents the end of calculating time.

Considering time delay exists in patents citation, and number of patents citation is relatively small in recently years, therefore, the end of 2015 is chosen as the searching time, specifically December 10, 2015. This paper retrieves 7272 patents related to wireless charging after searching the USPTO’s Issued Patents, involving a total of 91936 cited patents, and the searching time was December 10, 2015.After removing patents missing partial field, zero-cited patents and patents consolidating ‘one country and more yards’, a research data set of 240 valid sample patents was chosen, including 30031 cited patents and 91936 citations.

to a certain extent[11]. At the same time, the span also presents the value of a patent. Through the analysis of the field of wireless charging from the perspective of patent life span, five conclusions can be drawn as follows:

在基层实验室人员建设方面，相关部门要做好相关的技术人员招聘工作。作为基层部门更需要有专业的技术人员加强对实验操作的管理。对于上岗的工作人员要进行上岗前的技术培训工作，培训完后具备上岗能力的人员要给予一定的认可证，让工作人员能够持证上岗。另外，在实验操作方面，要加强工作人员的安全管理，构建实验操作平台供工作人员进行学习，及时补充自身的不足，给实验操作带来一定性的发展。

The core ideas of ‘double-proportion’ obsolescence method are as follows. First, given a patent in the sample set, T is defined as the first consecutive period (T units) that the patent received no citation, P1 is defined as the proportion of the number of citations the patent accrued till the end of T to the total number of citations the patent received during the entire observation window, and P2 is defined as the proportion of the number of patents meeting P1 and T constraints to the total number of patents in the sample set. Then T is iterated from small to large values to find the minimal T value that meets the given double-proportion P1 and P2 values as the optimal T, which will be used to determine the expiration time of the sample patent set and the judging criteria of its life length. The core steps of ‘double-proportion’ obsolescence method are as follows[8]:

(3) Calculate the cumulative citation numbers of the ith patent from the authorized time to each period of time j as the total citation numbers of the ith patent in statistical period as the percentage of the cumulative citation numbers of the ith patent from the authorized time to each period of time j as

大梦初醒后，我发现自己于叶子而言不过是可有可无的角色。维系我们之间的那根线太细了，如果我再不攥紧，它将会随时断裂。时机已容不得再酝酿，食堂里，像是一场自然而然的偶遇，我端着餐盘坐在了叶子的对面，把一盘土豆丝一根根吃得快见了底，才终于开了口，“你喜欢什么样的男生？”

(4) Count the starting time of the first consecutive time (T units) without citations of the ith patent as ti. The percentage of cumulative citation numbers of the ith patent before ti is Pi=Nij/Si. If Pi ≥ P1, and ti will be considered as the obsolescence point of the ith patent.

(5) Count the proportion of the patents that meet condition P1 in the sample patents as P’. If P’ first meets the condition P’≥P2 when T takes a certain value, T will be considered as the determine criteria of the obsolescence point of this sample patent set. The life span of the ith patent will be the temporal difference between ti and its authorized time.

2　Measurement of patent life span in the field of wireless charging

2.1　Datasetacquisition

90年代至今，潮州麦秆画经过不断地尝试与创新，以崭新的艺术面貌与艺术感染力在国内外工艺作品展上屡获佳绩，为潮州优秀的民间工艺美术增添了浓墨重彩的一笔。潮州麦秆画在2009年已列入广东省非物质文化遗产名录，领军人物方志伟在2016年被评为国家级工艺美术师，潮州麦秆画所拥有的艺术价值得我们深入探索与研究。

2.2　Data set preprocessing

Based on the relationship between patents in terms of citation, starting from the patents authorization, citation time, the citation frequency within that month and total citing frequency of each patent is calculated by month.The specific calculation method is shown in Table 1, which lists the timing distribution of citation in the patent No. 5923544t (document ‘Noncontact power transmitting apparatus’) in detail. The patent number ‘5923544’ is a patent granted to Japan’s TDK Corporation in July 1999, therefore there was no citation before July 1999. It can be seen from the table that up to December 2015, the patent received a total of 84 citations. It was first cited in the 7th month after obtaining authorization, then second citation occurred in the 11th month, and it was cited again in the 14th month. Afterwards, the patent received attention continuously. Up to January 2013, the cumulative number of citations was 42 and the cumulative citation proportion reached 50% for the first time. The longest consecutive continuous period of zero citation was 17 months. The span of the patent from the first citation to the statistical deadline was 165 months.

2.2.3 人群分布情况 男性发病5 875例，占发病总数的90.82%，女性发病594例，占9.18%，男女性别比为9.89∶1。职业分布，农民为高风险人群，发病6 166例，占总病例数的95.32%，其次是学生75例，占1.16%，最低是幼托儿童7例，0.11%。年龄分布以20～44岁组青壮年发病为主，4 879例，占75.42%，60岁及以上组发病最低，89例，占1.38%，见表2。

Table 1　Patent 5923544 citation sequence distribution statistics table

CitationtimeCitationfrequencyCumulativecitationfrequencyCumulativepercentCitationtimeCitationfrequencyCumulativecitationfrequencyCumulativepercent…197601000．00002010101370．4405…000．0000…0370．4405200002110．01192011051380．4524…010．0119…0380．4524200006120．02382012021390．4643…020．02382012031400．4762200009130．0357…0400．4762…030．03572012071410．4881200012140．0476…0410．4881200103150．05952013011420．5000…050．05952013020420．5000200107160．0714……………060．07142015071750．8929200111170．08332015084790．9405…070．08332015092810．9643200208180．09522015102830．9881…………2015111841．0000

2.3　PatentcitationdelayDE

进入新世纪之后，面对世情、国情、党情新变化的形势，江泽民于2001年7月在纪念建党80周年大会上的讲话全面阐述了“三个代表”重要思想。讲话明确指出：我们党必须始终代表中国先进生产力的发展要求，代表中国先进文化的前进方向，代表中国最广大人民的根本利益。这是我们党的立党之本，执政之基，力量之源。

It is found that all sample patents have different lengths of citation delay. Among the sample patents studied in this paper, the patent with the longest time delay is ‘5717311’ granted to Japan’s Ulead Corporation in February 1998. This patent was first cited in November 2014 and the citation delay is 201 months; the patent with the shortest citation delay is patent ‘8588689’ authorized to Microsoft Corporation in November 2013, and it was first cited in December 2013. The citation delay is only 1 month. It can be found that there is a great difference between citation delays among different patents. In order to reduce the influence of the citation delay on the patent life span, this paper calculates the mean DE of citation delay of sample patent and applies it to the judgment of ‘double-proportion’ obsolescence. Through the calculation, the average citation delay for sample patents is 27 months.

2.4　Calculation of patent T value

The criteria to determine patent T value is that a patent has not been cited in consecutive time(T units) in certain period starting from the patent’s authorization to the end of the observation period (December 2015). With the introduction of ‘citation delay’, patents that require value T to be determined are divided into three categories. In the first category, if a patent has not been cited within the (T+DE) unit time after the authorization, it is considered to be obsolete with no length of life since its life has not yet begun. Such patents cannot be included in subsequent studies. In the second category, if a patent is cited within the (T+DE) unit time after the authorization, the time of first citation is the starting point, after that, if the patent has not been cited for consecutive time (T unit) during the observation period, it has an observable length of life. Therefore the time of obsolescence is defined as the starting point of consecutive time phrases of no citation in the first T unit time, and the length of life refers to the time period from the authorization date to this starting point. In the third category, if a patent is cited within the (T+DE) unit time after the authorization, and there is no case of ‘no citation within consecutive T unit time’ till the end of observation period, which means the patent has obsolescence time and length of life and has been frequently cited during the observation period. The actual time of obsolescence cannot be observed because patent’s life has not yet ended till the end of the observation period (December 2015). Such patent is known as the right-censored patent.

To determine the proportional P, most of the scholars are in favor of 80-20 rule in their researches, and they believe that when a paper or a webpage has been browsed by 80 percent of users, even if there will be other users to browse continuously, the value of the information at that time has been greatly reduced. Compared with the paper or webpage information, the patent information has worse timeliness. Meanwhile, as an emerging technology, patents with serious right-censored phenomenon exist in the field of wireless charging technology , so in this paper, obsolescence proportion P1 and P2 are set as 60%, and this is set as a criterion to determine the validity of T.

This paper employs exhaustive method to determine the validity of T. Combined with the research and common sense of scholars, this paper sets the range of T from 1 to 50. Under the condition of double-proportion 60% -60%, the obsolescence of the sample patent was calculated when the value of T is taken as 1, 2... 50 respectively. Table 2 shows the distribution and obsolescence rates of selected patent categories under different T, the observation time is from January 1995 to December 2015, and the total number of sample patents is 4541. The first type of patents in the table refers to the patent that hasn’t been cited for the first time in the period of (T + DE) months from the approval date; the second type of patents refer to the number of obsolescence patents during the observation time; the third type of patents refers to the number of patents that are not expired during the observation time;the number of obsolescence P1 patent refers to the number of invalid patent citation within the effective life span occupying more than P1 (60%) of the total citation number during observation time; the obsolescence-P2 proportion refers to the proportion of the number of invalid-P1 patents and the number of non-obsolescence patents in the total number of invalid patents and non-obsolescence patents.

Table 2　T value determination statistics excerpt

TThefirsttypeofpatentNumberProportionThesecondtypeofpatentNumberProportionThethirdtypeofpatentNumberProportionValid⁃P1patentnumberValid⁃P2proportion115740．34729240．644430．0095070．185215000．33029310．6451100．0245530．218………………………128960．19723610．52012840．2835960．516138510．18723080．50813820．3046080．539148080．17822340．49214990．3306070．564157660．16921590．47516160．3566130．590167300．16120810．45817300．3816050．613176990．15419910．43818510．4085960．637186640．14619250．42419520．4305820．654196370．14018630．41020410．4495750．670206110．13518020．39721280．4695750．688………………………

From the statistical results, it can be seen that when T is 16, the proportion of patents with citation frequency of transfer obsolescence greater than P1 is 61.3%, exceeding P2 (60%) for the first time, at the same time, it can be found that when T is equal to 16, the total citation frequency of obsolescence patent and the number of obsolescence-P1 patent begins to show a downward trend. Thus, under the criterion of double-proportion 60%-60%, when T is equal to 16, the judgment condition is satisfied and the total cited frequency of obsolescence patent reaches the maximum under the condition of double-proportion, so effective T is 16.

2.5　Analysis of patent life span

Before 1995, wireless charging related technologies were at an embryonic stage, lacking the basis for snowball marketing due to its low user attention. In this stage, 6475 patents authorized from 1995 to 2015 are analyzed, of which altogether 4541 patents are cited during the observation time period. The life span of 4541 sample patents are measured based on the standard that effective T value is 16. Among them, 730 sample patents had not been cited for 43 consecutive months (DE+T) since authorization, 2081 patents had a specific obsolescence point during observation time, and 1730 patents had not obsolesced during observation time, that is, the right-censored patents. Table 3 lists the life span of 2081 obsolesced patents during observing time.

Table 3　Life span distribution of sample patents

Patentlifespan(month)PatentquantityPatentquantityproportionPatentlifespan(month)PatentquantityPatentquantityproportion[0-36)129262．09%[100-120)200．96%[36-60)55626．72%[120-200)291．39%[60-100)1848．84%///

The shortest patent life span of obsolesced patents is 4 months, and the longest one is 196 months. Over 80% of the patents have a life span of 60 months (5 years). Only 1.39% of patents have a life span over 10 years, and these patents have strong vitality as well as great influence on the development of wireless charging technology. It can be found that 98.61% of patents complete their patent value within 120 months (10 years), and lose their dissemination and utilization value. In general, patent life span in the field of wireless charging is overall shorter, the obsolescence rate is faster, and related technology is still in a rapid development stage (Fig.1).

Fig.1　Life span distribution of sample patents

According to the table of patent life span and survival analysis function, further analysis on the survival of the sample patents is conducted. As shown in Table 4, one month is set as unit interval span, patent life span is divided into 247 time segments, the monthly life span distribution of sample patents is counted respectively and finally the relevant life function images can be drawn. The life function images include curves of the cumulative survival function, the density function and the hazard function, respectively.

The curve of cumulative survival rate reflects the proportion of the number of patents that is still active in the total number of samples per time unit. It can be seen that patents in the field of wireless charging decline significantly in the first 6 years (Fig.2). The patents in this field have a generally low survival rate, and life span is mostly within 6 years; the probability density function reflects the distribution rule of the sample patents in each life span interval, namely the proportion of the patents with same life span. As indicated in Fig.3, the probability density concentrates in the first 6 years life span intervals, illustrating that the first 6 years after the authorization is the core time for patents to play their value. Two patent obsolescence peaks lie in the third year and the fifth year after authorization and patents in this field that last until 8 years later are of less number; the hazard function reflects the probability of patent obsolescence in each life span interval. As shown in Fig.4, the curve has several rises and falls. In the 4 time segments of the early period, the probability of patent obsolescence is larger and it indicates a trend of rapid growth. After that the curve rapidly drops and the obsolescence probability declines to a low level showing a gentle trend of change. From here we see that the patent faces relatively high obsolescence risk in the first 5 years after authorization, and then the patent hazard rate begins to decline and in the eighth and eleventh year reaches a small peak of obsolescence. The obsolescence probability of the sample patent that has gone through the obsolescence peak becomes smaller and it is cited continuously by other patents later on, and finally becomes the core technology patent in the field.

Table 4　Life table of patent information (excerpt)

IntervaltimeEnterquantity(ei)Limitedquantity(li)Obsolescencequantity(di)Adventurequantity(ni)Conditionalobsolescencerate(qi)Conditionalsurvivalrate(pi)Cumulativesurvivalrate(si)Probabilitydensity(fi)Hazardrate(hi)138110038110．00%100%100%0．00%0．00%238110038110．00%100%100%0．00%0．00%338110038110．00%100%100%0．00%0．00%438110038110．00%100%100%0．00%0．00%538110538110．13%99．87%100%0．00%0．13%…………………………21297431482958．51．62%98．38%84．06%1．40%1．64%222895165528871．91%98．09%82．69%1．36%1．92%232824205328141．88%98．12%81．12%1．58%1．90%24275131492735．51．79%98．21%79．59%1．53%1．81%25267129372656．51．39%98．61%78．16%1．43%1．40%…………………………2456105．50．00%100．00%16．80%0．00%0．00%24652040．00%100．00%16．80%0．00%0．00%24730030．00%100．00%16．80%0．00%0．00%

Fig.2　The distribution map of cumulative survival rate

Fig.3　The distribution map of probability density

Fig.4　The distribution map of hazard rate

3　Conclusion

The changes of patent information life span present the development stage of the related technical field

(1) Determine N pieces of patents, and let Ni represent the ith piece of patent.

From the perspective of the analysis method, the paper improves‘double-proportion’ obsolescence method according to the particularity of patent information, and reduces the double-proportional P value to conform to the characteristics of patent information. However, as an emerging technology[12], wireless charging is at the peak of patent application in recent years. The time scale selected in this paper ended by the end of the observation, hence the data in recent years cannot be acquired completely, and a certain number of ‘right-censored’ data has not been included, which may result in system errors.

2.4.3 共有峰的指认及相关分析 根据MS离子峰及碎片信息与保留时间，综合对照品MS信息及文献报道的MS数据[11-14]对药材样品的指纹图谱中的共有峰进行结构归属的初步判断。通过与混合对照品MS图谱的离子峰信息进行比对，确定峰4、8、9、11、14、24、25分别是没食子酸、氧化芍药苷、没食子酸甲酯、丹皮酚原苷、芍药苷、苯甲酰氧化芍药苷、丹皮酚；推测鉴定出了29个共有峰，详见表4。

From the perspective of the analysis results, an average citation delay of patents in the field of wireless charging is 27 months. On one hand, because wireless charging has just got rapid development in recent two years, the early patents in the retrieval results have weak connection with wireless charging, and the phenomenon of innovation based on the early patents is common. On the other hand, recent patents were partially counted due to the time limits for data acquisition, leading to a slightly larger citation delay.

From the perspective of the patent T value, patents in the field of wireless charging are obsolesced if they are not cited for consecutive 16 months. This emphasizes the timeliness of patents in this field compared with other industries, and also shows that the field is still in the rapid developing stage with a rapid update and iteration speed of the patents[13]. A large number of low-value patents have been quickly eliminated.

2.4 疾病隐患 岸边水陆交叉污染带容易滋生各种病菌和病原体，加上消落区域泥沙淤积，使得病菌难以扩散，容易诱发各类流行疾病，危害动植物的健康和生长。

From the perspective of the average patent life span, 60% of the patents reach the obsolescence point within a year and a half while nearly 90% of the patents will reach the end of life within 5 years. Fewer patents have long-term influence on the development of wireless charging technology. It generally shows a feature that the life span is short while the eliminating speed is fast.

From the curve of patent survival analysis, it can be seen that the first 1-5 years is the peak period of patent obsolescence, in which the obsolescence patent quantity has the widest distribution and the obsolescence risk is at the greatest level. In the early stage of development, a large number of patentees applied

for massive patents to preempt the market, resulting in the lack of core technology in patents and inability to keep a long-term development. Referring to the conclusions above, the future inventors should consider the characteristics of patents in the field of wireless charging, abstract the technological key points from a large number of new authorized patents, pay attention to information integration and increase investment, thereby contributing more to patents with strong vitality.

This study adopts the ‘double-proportion’ obsolescence method to analyze the patent life span in the field of wireless charging. Because wireless charging is a newly developing field with a short developing time, the retrieval, acquisition and statistics of patent information is still incomplete. The field with mature technology will be selected for further research. In addition, the selection ranges of samples still need further extension. The USPTO data selected in this paper limits the national scope of patent applications in the field of wireless charging, and the imperfection of samples may also bring deviation to the measure of life span; at the same time, due to the limited space, the core enterprises in the field of wireless charging have not been included for technology competitiveness evaluation. Thus, the evaluation of the competitiveness and development prospects of related companies can be evaluated in future researches by investigating the life span of patents they apply.

References

[ 1] Harter S P, Kim H J. Accessing electronic journals and other e-publications: an empirical study [J]. College & ResearchLibraries, 1996, 57(5)：440-456

[ 2] Bugeja M, Dimitrova D V. The half -life phenomenon: eroding citations in journals[J]. SerialsLibrarian, 2006, 49 (3)：115-123

[ 3] Koehler W. An analysis of Web page and Web site constancy and permanence[J]. JournalofAmericanSocietyforInformationScience, 1999, 50(2)：162-180

[ 4] Wang Y, Liu Y Q, Zhang M, et al. Modeling lifetime of web pages based on user interest analysis[J]. JournalofChineseInformationProcessing, 2008, 22(2)：76-80

[ 5] Gay C, Lebas C, Patel P, et al. The determinants of patent citations: an empirical analysis of French and British patents in the US[J]. EconomicsofInnovationandNewTechnology, 2005,14(5)：339-350

[ 6] Patel D, Ward M R. Using patent citation patterns to infer innovation market competition[J]. ResearchPolicy, 2011, 40(6)：886-894

[ 7] Wang J C. New understanding of information obsolescence: the generation and attenuation of information value[J]. JournalofTheChinaSocietyforScientific,2013,32(4)：354-362 (In Chinese)

[ 8] Zhu M X, Xu H X, Gao J. A comparative study of half-life of network information resources with different contents[J]. JournalofIntelligence, 2010, 29(9)：29-32

[ 9] Wang J C,Yang K, Xiang Li, et al. Obsolescence determination of network information based on ‘double-proportion’ method[J]. ChineseJournalofLibraryandInformationScience,2013, 6(4)：28-39 (In Chinese)

[10] Wang Y D,Hu Y F,Hu D.Research on the ‘citing time lag’ phenomenon in journal papers[J]. IndormationStudies: Theory & Application,2015, 38(8)：56-60

[11] Narin N, Franci S, Ellio T,et al.Patents as indicators of corporate technological strength[J].ResearchPolicy,1987,16(2-4)：143-155

[12] Cho T S, Shih H Y. Patent citation network analysis of core and emerging technologies in Taiwan: 1997-2008[J]. Scientometrics, 2011, 89(3)：795-811

[13] Duguet E, Macgarvie M. How well do patent citations measure flows of technologies? Evidence from French innovative surveys[J]. EconomicsofInnovationandNewTechnology,2004,14(5)：375-393

ShiMin, born in 1974. She is a Ph.D candidate of Information Management School at Wuhan University. Meanwhile, she is an associate researcher serving in Institute of Scientific and Technical Information of Guangxi .She was a visiting scholar of Oxford University from November 2015 to November 2016. Her current research interests focus on technology transfer services research, Intellectual property research, scientific and technical finance research.