外文翻譯--關(guān)于大數(shù)據(jù)那些事

1、<p>　　Inventory: Grilled steak that thing big data</p><p>　　Today, industry and academia have been discussing a word, and that is big data. Whether or IT circles academic circles, as long as something t

2、o talk about big data becomes very big on. However, a large data mining , big data analytics, big data marketing, etc. just the beginning of things, for most companies, big data is still a strong mystery. Thus, we do not

3、 yet fully understand how to use big data mining, the deification too large variety of data has heard the opinion of course, there are a l</p><p>　　Then, standing on an objective point of view, around the fo

4、llowing questions to share with you a few points about the big data, also Pa Pa those things big data: </p><p>　　A large marketing and disclosure of personal privacy data whether with or without cause and lo

5、gic? </p><p>　　2, big data marketing can bring in the end what kind of enterprise value? What is the value in the end user can bring? Are user negate or objectionable big data marketing? </p><p>

6、;　　3, how to properly treat big data? How do relations and traditional survey methods for large data or statistics? </p><p>　　4, big data marketing exactly what challenges facing? </p><p>　　A ra

7、pid development and data privacy concerns of big data attendant</p><p>　　The emergence of social media, so that the number of users to share data, to the extent incalculable. And now, the type of social medi

8、a continues unabated, the greater the popularity of smart phones, letting more users shift to the mobile Internet, which in turn further contribute more data and content. Such data increment to the global social media re

9、venue rose, according to consulting firm only Gartner2012 year results show that in 2012 the global social media revenue is estimated to reach $ 1</p><p>　　While social media is because large pots full of da

10、ta, the other is the user constantly unreserved personal information to the Internet, the information including age, gender, geography, living conditions, attitudes, whereabouts, hobbies, consumer behavior, health status

11、 or even sexual orientation. For a time, for large data mining massive user information, big data analytics, big data precision marketing, advertising and accurate delivery, etc. quickly put on the agenda of major compan

12、ies. </p><p>　　For example, a true story happened in the United States will tell us how to master the use of data mining, our whereabouts. An American family has received delivery of a shopping mall on mater

13、nity promotional coupons, promotional coupons obvious who is to give the home 16-year-old girl. The girl's father was angry and looking for the mall to discuss that. But a few days later, the father found that 16-yea

14、r-old daughter really was pregnant. The reason why the prophetic mall, it is through a large</p><p>　　After similar large data mining and marketing events occurred more today, especially social media to gene

15、rate large amounts of data. As a result, many people began to worry about data privacy, began to criticize big data precision marketing violated privacy concerns we enter the era of big data out of control, and the reaso

16、n is due to more social media. </p><p>　　Two, not completely draw large data between marketing and personal privacy disclosure equals! Logic does not hold! </p><p>　　If an objective analysis of

17、the above problems will find that this is a difficult Takahashi the chicken or the egg problem. Blindly criticize big data analysis of personal user data leakage or misuse is not objective. </p><p>　　Because

18、 the nature of social media is to share and spread the emergence of social media really meet people to share personal information, drying various data desires, so that people living in the past suddenly silent transferre

19、d to let the whole world see themselves the platform. It thus reached the inner satisfaction and sense of place. So, just from the individual psychology behind to consider social media is beneficial for them, they do not

20、 think they are hidden secret contribution, Now share</p><p>　　Of course, if the social media platform casual user misuse or disclosure of background data, such as personal contact details, home address, ban

21、k and other highly secret information, which is indeed a naked violation of privacy, there is no great moral, must be condemned and the law sanctions. </p><p>　　So, from this perspective, the big data precis

22、ion marketing and personal initiative to share and disseminate information and data on the network between no contradiction at first people might be surprised: Why They know what I want to know why they are my needs? ? B

23、ut with 'discern the mind' push behavior make people's lives more convenient time, such as eliminating the need for a lot of searching, finding and comparing the time a product or service, they may be very ac

24、customed to and rely on su</p><p>　　Therefore, users publish and share information about whether privacy, before users can share information and screening done careful consideration. It is very important tha

25、t this is an invasion of privacy boundaries or not. Those are the user chooses to not fit or do not want to publish others know the information that users think privacy, and those who have been publicly posted on social

26、media or network user information were considered to be spread. </p><p>　　Therefore, the general public for the mass analysis of information, mining, classification, and thus precision marketing of big data

27、behavior can not be criticized as being blind to harm the interests of users, and those users are stored in a certain position, do not want to be someone else understanding of the information (personal information stored

28、) If the disclosure or use of people with ulterior motives, and that this is the violation of privacy. But this can not be blamed for big data, and</p><p>　　Therefore, we can not interpret large data precisi

29、on marketing too. In fact, the nature of the problem is that people really care about the fate of the clutter of information (related to the sharing of information and motivation behind psychological)? And big data marke

30、ting really touched people hidden secret or bottom (requires redefinition of secrets and the bottom line)? Because, if people are open to share the default, then the concept of invasion of privacy is not established. If

31、people do </p><p>　　| Three, big data marketing business and give users exactly what is the value? </p><p>　　After having discussed the above issues, we should treat the data precision marketing

32、 sincere about it? So what big data marketing for both businesses and consumers, it has what value? </p><p>　　| 1, the value for the enterprise </p><p>　　Let us look at a foreign case: </p>

33、;<p>　　We all know that the US drama << >> house of cards, mention << >> successful house of cards, the maximum credit is the major data analysis. Therefore, >> << house of card

34、s has become almost a classic case of big data marketing, but also Netflix, the US-based mining user information to determine the content of the production of a successful attempt. </p><p>　　Then we look at

35、a domestic case: </p><p>　　We all know that Ali Baba and the Sina microblogging cooperation things to spend $ 586 million stake in Alibaba, Sina Weibo. In addition to the analysis of the major media networks

36、 that Alibaba hopes to build ecosystem, strengthen the flow inlet, Tencent and other reasons the challenge In addition, there is an important reason is probably the big data marketing strategy. </p><p>　　Tod

37、ay, major Internet bigwigs are staking their claims, trapping the user, who circled the user, allowing users active on its platform, the user grasps the wealth of information (including significant information in the for

38、eground and background information hiding ) Sina Weibo in China has hundreds of millions of users, this volume is very large, but if you can not put these Sina user-generated information and reasonable use, it is a huge

39、waste of those resources. We look Alibaba, China's larges</p><p>　　| 2, the value for the user </p><p>　　These two examples are said to bring the value of big data business, then marketing f

40、or big data users, in the end there is no value if the user is very offensive precision marketing let us look at a new survey data??: </p><p>　　National Institute of Advertising Communication University of C

41、hina has just released a 2014 Sino << >> Mobile Internet Development Report, the survey compared the Sino-US users of mobile Internet usage, as well as mobile user attitudes toward mobile advertising. </p&

42、gt;<p>　　Survey shows that most likely to be intelligent end-user response advertising content is: (1) the user to purchase items related advertising (2) To purchase items associated with the coupon (3) funny ad (

43、4) with the user's favorite brand-related advertising (5) with a user visited a website or online application used relevant ads (6) and the recent online shopping related ads (7) places relevant ads with the user'

44、;s (8) listen and recently, watch the broadcast / TV-related advertising. (accountin</p><p>　　From these data, we can see the results in the eight, six are linked with large data precision marketing relation

45、ships. For example, the user wants to purchase items related ads, arouse the user's response or interaction. How understand? premise marketing is big data computing and speculated that the real needs of users, to see

46、 what users need to buy related products, and then give the user direct push users want, like, so that precise distance. So users? users willing to such advertising or pro</p><p>　　Therefore, this result sho

47、ws that big data is not entirely accurate marketing will allow users dislike, but to see the extent of your mind to discern the user. Therefore, if you push the contents of the article and the user wants to buy correlate

48、d with most users favorite brand-related, etc. then this precise digging and users will not be offensive, but give users convenience. </p><p>　　| Fourth, do not blindly believe in big data, the essence of wh

49、at is big data? </p><p>　　Read the above analysis, perhaps you will think of anything big data analysis. However, we can not blindly believe in big data, so the next question arises. </p><p>　　|

50、 1, what kind of relationship analysis of big data and traditional statistical methods? </p><p>　　Large data compliance is: a lot of data, and even all the data, and then use the algorithm to calculate the a

51、nalysis to find a more accurate correlation (not causation) between the various factors in order to discover the laws of data. </p><p>　　Then we look at the traditional statistical methods, statistical analy

52、sis is to learn how to solve by choosing a small number of samples, through sample analysis, and then infer the overall trends and patterns. So, with the probability of 90% will normally be prescribed , the maximum exten

53、t of 95% or 98% confidence level (precision) of the overall estimation. If the object clearly, the sample is selected properly, the operation of science, it does not need to be able to analyze the law of large </p>

54、<p>　　To give an example of inappropriate for understanding: Suppose selected sample of 1000, concluded that the law is A, select the 2000 sample, showing A similar law, then select 3000 almost so, we actually cho

55、ose science over 1000 samples. we can achieve the purpose. Therefore, the traditional sampling and statistical methods, to the maximum extent, solve the cost problem, although there will be errors, but still can be found

56、 in the law explicitly. </p><p>　　So, from this perspective, the big data analysis is likely to end up with the traditional statistical methods similar results analysis, only a small sample of the original i

57、nto a large sample analysis. While big data analysis is theoretically more accurate also can make up for deficiencies of traditional error, but accuracy is not like that very much improve our imagination (because of the

58、large data analysis will be severely affected by the data source). In addition, not be able to find more ne</p><p>　　In addition, among the traditional statistical analysis, such as analysis of the market si

59、tuation, we must be practical environment and background to interpret the data and analyze the data, we do not regard the data as a unique and universal guidelines. So, there is the presence of human analysis of the proc

60、ess data based on experience and the actual situation, while the ability to participate in the analysis is very important. </p><p>　　| 2, what kind of things can not do big data, and the traditional method o

61、f investigation and analysis, but can be done? </p><p>　　Big data marketing premise is the big data analysis, and big data analysis is based on the algorithm, a computer solidification mode. That is, the ori

62、ginal part of the work that the people of the data analysis, and now we agreed to put it in the algorithm. Also, the large data precision marketing is a web browser for user-generated data, share data, search data, and s

63、o the behavior of information for analysis, thus classifying people or things, and thus speculated that human preferences, interes</p><p>　　However, preference is not equal to the real needs, like a person d

64、oes not necessarily mean clicking on social media today, said:. 'The good products', they think he must have liked or necessarily need this product? </p><p>　　The behavior of the machine can be class

65、ified, but it can not really detect the psychological and the real needs of people. So, for the detection of human psychology and the real needs, how do we do? At this time, the traditional market research and analysis m

66、ethods are irreplaceable For example, in-depth interview, such as focus group interviews, the projection method, etc. These methods are the maximum extent possible, from a psychological point of view to analyze and find

67、that people really d</p><p>　　From this perspective, big data is not a panacea, can not be blindly myth, we must clearly recognize its essence, it can be used to do, not for what we can understand: people ca

68、lculation of the data and analysis may be replaced by a machine now, but another part of people's work (the ability to detect the human heart) algorithm can not be replaced. </p><p>　　For example, two ye

69、ars ago to write a book, I had reported << algorithm can be used to automate, and what means to save publishing >> this new technology, a large number of books on Amazon.com allegedly are currently being writ

70、ten out of the algorithm, the algorithm based on People write books logical thinking to organize language. However, these books can not compensate for the lack of human emotions, can not express the social context and th

71、e environment of emotional upheaval brought more. </p><p>　　| Five, the real challenge for large data analysis or big data marketing facing what is? </p><p>　　| 1, data redundancy, there is no n

72、eed to use so much data? </p><p>　　Data Sources, data quality or without security, whether it is really needed? </p><p>　　Big Data analysis has been lauded advantage is:??? The use of massive am

73、ounts of data, but the data is not better how to filter the data to find valuable and useful data and redundant data would be huge for large data analyze what kind of impact? </p><p>　　For large data, the da

74、ta source is to analyze the accuracy of massive fundamental guarantee, however, the amount of data to a certain extent also facing a big problem: it wants to ensure the accuracy of the difficulties of such a change would

75、 be difficult to guarantee The accuracy of the analysis results. Examples of big data analysis and prediction of failure, there are many. For example, one of the most typical and famous is Google Flu Trends predicted fai

76、lure cases. </p><p>　　Reported that Google is a search engine based on the analysis of the data, its analysis and monitoring data from the US Centers for Disease Prevention and Control difference of nearly t

77、wice as much. Although Google constantly adjust the algorithm, but still can not guarantee the accuracy of the results. This indicates an important Problem: Data Sources Google search engine is based on the search term u

78、sed to analyze many search terms are invalid, does not make any sense, so they do not really re</p><p>　　So you get these data, how do you guarantee that they needed? Really important? If a serious deviation

79、 data source, then your analysis and then precise, so futile. For example, you spend a lot of energy sharing information on the Internet daily to collect user-generated, you have all the information they analyzed the res

80、ults of several consumer trends predicted. However, the sharing of information in a large number of redundant information, data accuracy is poor, many of which are now Consumptio</p><p>　　| 2, Gangster platf

81、orm game, general business difficult to master large amounts of data, it is difficult test credibility </p><p>　　The company holds major Internet platform user resources, user-generated information is gather

82、ed in the course of each platform. However, companies or data platform and will not be fully open to the public. We can only crawl to pass certain tools scattered information on the network, but can not accurately grasp

83、the full significance of practical value and background data and information. </p><p>　　These vast amounts of information, such as Google for such big Internet companies, is the treasure of Big Data may be j

84、ust these bigwigs platform game, ordinary businesses are more difficult to get involved. </p><p>　　Also, do not open interoperability between these platforms and their results are not out of the data analysi

85、s of third-party validation and testing, we will not know the results of their analysis of large data validity and credibility. Of course, they will be those Data analysis on the user's own product development and it

86、s development is still very valuable. So, ordinary or common enterprise desire for big data might be wishful thinking. future large Internet platform company might sell big dat</p><p>　　In addition, a large

87、data analysis algorithms currently no standard, there is no generally accepted and unified and effective tool. </p><p>　　So, from these aspects, big data analytics and big data marketing is still a long way

88、to go before we need to correct, rational view of big data (source:.. IDonews, compile: Free Paper Download Center </p><p>　　盤點：扒扒大數(shù)據(jù)的那些事兒</p><p>　　如今，業(yè)界和學(xué)術(shù)界一直在討論一個詞，那就是大數(shù)據(jù)。不管是學(xué)術(shù)圈還是IT圈，只要能談?wù)擖c兒大數(shù)

89、據(jù)就顯得很高大上。然而，大數(shù)據(jù)挖掘、大數(shù)據(jù)分析、大數(shù)據(jù)營銷等等事情僅僅只是個開始，對大多數(shù)公司來說，大數(shù)據(jù)仍有很強的神秘色彩。于是，在我們還沒有完全搞明白如何運用大數(shù)據(jù)進行挖掘時，各種過于神化大數(shù)據(jù)的輿論就已經(jīng)不絕于耳了。當(dāng)然，也有很多人直接批判大數(shù)據(jù)或大數(shù)據(jù)營銷給我們造成的隱私威脅。也有很多人根本沒有搞清楚什么是大數(shù)據(jù)，到底有什么價值。</p><p>　　于是，站在客觀的角度，圍繞下面幾個問題與大家分享有關(guān)大

90、數(shù)據(jù)的幾個觀點，也扒扒大數(shù)據(jù)的那些事兒：</p><p>　　1、大數(shù)據(jù)營銷和個人隱私泄露究竟有無因果和邏輯關(guān)系？</p><p>　　2、大數(shù)據(jù)營銷到底能帶給企業(yè)什么樣的價值？到底能帶給用戶什么價值？用戶是否全盤否定或反感大數(shù)據(jù)營銷？</p><p>　　3、如何正確看待大數(shù)據(jù)？如何看待大數(shù)據(jù)和傳統(tǒng)調(diào)查方法或統(tǒng)計學(xué)的關(guān)系？</p><p>

91、　　4、大數(shù)據(jù)營銷究竟面臨什么樣的挑戰(zhàn)？</p><p>　　|一、大數(shù)據(jù)的迅猛發(fā)展與數(shù)據(jù)隱私的憂慮相伴而生</p><p>　　社交媒體的出現(xiàn)，讓用戶數(shù)據(jù)的分享數(shù)量達到了難以估量的程度。而如今，社交媒體的種類有增無減，智能手機的更大普及，又讓更多用戶轉(zhuǎn)移到移動互聯(lián)網(wǎng)，從而又進一步貢獻更多數(shù)據(jù)和內(nèi)容。這樣的數(shù)據(jù)增量讓全球社交媒體的收入大漲，僅根據(jù)咨詢公司Gartner2012年的研究結(jié)果顯

92、示，2012年全球社交媒體收入估計達到169億美元。</p><p>　　一邊是社交媒體因為大數(shù)據(jù)的盆缽滿載，另一方面則是用戶不斷毫無保留的將個人信息交給互聯(lián)網(wǎng)，這些信息包括年齡、性別、地域、生活狀態(tài)、態(tài)度、行蹤、興趣愛好、消費行為、健康狀況甚至是性取向等。一時間，針對海量用戶信息的大數(shù)據(jù)挖掘、大數(shù)據(jù)分析、大數(shù)據(jù)精準(zhǔn)營銷、廣告精準(zhǔn)投放等等迅速被各大公司提上日程。</p><p>　　比如，

93、一個發(fā)生在美國的真實故事就會告訴我們，利用數(shù)據(jù)挖掘如何掌握我們的行蹤。一個美國家庭收到了一家商場投送的關(guān)于孕婦用品的促銷劵，促銷劵很明顯是給給家中那位16歲女孩的。女孩的父親很生氣，并找商場討說法。但幾天后，這位父親發(fā)現(xiàn)，16歲的女兒真懷孕了。而商場之所以未卜先知，正是通過若干商品的大量消費數(shù)據(jù)來預(yù)估顧客的懷孕情況。</p><p>　　類似的大數(shù)據(jù)挖掘和營銷事件在今天更多的發(fā)生，尤其是社交媒體產(chǎn)生大量數(shù)據(jù)后。于

94、是，許多人對個人隱私數(shù)據(jù)開始擔(dān)憂，開始批判大數(shù)據(jù)精準(zhǔn)營銷侵犯了個人隱私，憂慮我們進入了大數(shù)據(jù)失控的時代，并將原因更多歸結(jié)于社交媒體。</p><p>　　|二、大數(shù)據(jù)營銷和個人隱私泄露之間不能完全劃等號！邏輯關(guān)系不成立！</p><p>　　如果客觀的分析一下上述問題就會發(fā)現(xiàn)，這是一個難以分說的雞生蛋還是蛋生雞的問題。一味地批判大數(shù)據(jù)分析對個人用戶數(shù)據(jù)的泄露或濫用是不客觀的。</p&

95、gt;<p>　　因為，社交媒體的本質(zhì)在于分享和傳播，社交媒體的出現(xiàn)的確滿足了人們分享個人信息、曬各種數(shù)據(jù)的欲望，讓人們在過去無聲無息的生活中突然轉(zhuǎn)移到了可以讓全世界看到自己的平臺上來。人們從而達到了內(nèi)心的滿足感和存在感。因此，單從個體的背后心理來考慮，社交媒體對他們來說是有益的，他們不認(rèn)為自己貢獻的是不可告人的秘密，既然分享出來，那一定是希望或允許別人看到的。因此，這是一種無形的默許的交易，用戶樂意把自己的各種瑣碎細(xì)節(jié)暴

96、露于社交媒體，而對社交媒體上雜亂無章的海量用戶數(shù)據(jù)進行有序的分類和分析也沒有什么不妥。</p><p>　　當(dāng)然，如果社交媒體平臺隨意濫用或泄露用戶的后臺數(shù)據(jù)，比如個人聯(lián)系方式、家庭住址、銀行等極為隱秘的信息，這的確是赤裸的侵犯隱私的行為，極其沒有道德，必須要受到譴責(zé)和法律制裁。</p><p>　　所以，從這個角度來看，大數(shù)據(jù)精準(zhǔn)營銷與個人主動分享和傳播到網(wǎng)絡(luò)上的信息數(shù)據(jù)之間并沒有矛盾。

97、人們起初或許會驚訝：為什么他們知道我想買什么？為什么他們知道我的需求？但隨著“猜透心思”的推送行為讓人們的生活越來越便利時，比如省去大量搜索、查找和對比產(chǎn)品或服務(wù)的時間，他們可能會十分習(xí)慣并依賴這種精準(zhǔn)性，并不會在意他們本來就隨意分享到網(wǎng)絡(luò)上的雜亂信息被如何挖掘和利用。</p><p>　　因此，用戶發(fā)布和分享的信息是否為隱私，在用戶分享信息之前就做過慎重考量和篩選。這一點非常重要，這是侵犯隱私與否的界限。那些被

98、用戶選擇為不適合發(fā)布或不希望別人知道的信息就是用戶認(rèn)為的隱私，而那些已經(jīng)公開發(fā)布到社交媒體或網(wǎng)絡(luò)上的信息則被用戶認(rèn)為是可以傳播的。</p><p>　　所以，普通的對海量公開信息的分析、挖掘、歸類，從而進行精準(zhǔn)營銷的大數(shù)據(jù)行為不能一味被罵成是對用戶利益的損害。而那些對用戶存儲在某些位置、不希望被他人了解的信息(私人存儲的信息)如果被別有用心的人泄露或利用，那這就是隱私侵犯行為。但這就不能歸罪于大數(shù)據(jù)，而應(yīng)質(zhì)問存貯