学年

教科

質問の種類

TOEIC・英語 大学生・専門学校生・社会人

この長文問題の答えと解説をお願いします。

15 語数: 398 語 出題校 法政大 5 We are already aware that our every move online is tracked and analyzed. But you 2-53 couldn't have known how much Facebook can learn about you from the smallest of social interactions - a 'like'*. (1) Researchers from the University of Cambridge designed (2) a simple machine-learning 2-54 system to predict Facebook users' personal information based solely on which pages they had liked. E "We were completely surprised by the accuracy of the predictions," says Michael 2-55 Kosinski, lead researcher of the project. Kosinski and colleagues built the system by scanning likes for a sample of 58,000 volunteers, and matching them up with other 10 profile details such as age, gender, and relationship status. They also matched up those likes with the results of personality and intelligence tests the volunteers had taken. The team then used their model to make predictions about other volunteers, based solely on their likes. The system can distinguish between the profiles of black and white Facebook users, 15 getting it right 95 percent of the time. It was also 90 percent accurate in separating males and females, Democrats and Republicans. Personality traits like openness and intelligence were also estimated based on likes, and were as accurate in some areas as a standard personality test designed for the task. Mixing what a user likes with many kinds of other data from their real-life activities could improve these predictions even more. 20 Voting records, utility bills and marriage records are already being added to Facebook's database, where they are easier to analyze. Facebook recently partnered with offline data companies, which all collect this kind of information. This move will allow even deeper insights into the behavior of the web users. 25 30 (3) - Sarah Downey, a lawyer and analyst with a privacy technology company, foresees insurers using the information gained by Facebook to help them identify risky customers, and perhaps charge them with higher fees. But there are potential benefits for users, too. Kosinski suggests that Facebook could end up as an online locker for your personal information, releasing your profiles at your command to help you with career planning. Downey says the research is the first solid example of the kinds of insights that can be made through Facebook. "This study is a great example of how the little things you do online show so much about you,” she says. "You might not remember liking things, " but Facebook remembers and (4) it all adds up.", * a 'like': フェイスブック上で個人の好みを表示する機能。 日本語版のフェイスブックでは「いいね!」 と表記される。 2-56 2-57 2-58 36

回答募集中 回答数: 0
数学 大学生・専門学校生・社会人

確率の勉強をしている学生なのですが、この問題が分かりません。どなたか教えていただけませんか。

練習問題 1.8 (積率母関数) X を非負の確率変数とし, x(t) = Eetx は全てのt∈ に対して有限であると仮定する.さらに,全てのt∈ R に対し E [XetX] < ∞ であると仮定する.この練習問題の目的は, '(t) = E [Xetx] で あり、特に'(0)=EX であることを示すことである。 微分の定義, すなわち次式を思い出そう. 4'(t) = lim x(t) - (s) lim st t-s st EetxEesx t-s 「etx = lim E st t-s 上式の極限は,連続な変数sについて取っているが,t に収束する実数列{8}n=1を 選ぶことができ, 次を計算すればよい. 「etx e³n X lim E sn→t t-Sn これは、次の確率変数の列 etx -enx Yn = t-Sn の期待値の極限を取っていることになる.もしこの極限が, t に収束する列{Sn}=1 の選び方によらず同じ値になるならば、この極限も limotE [ex と同じで,そ れは '(t) である. .tx sx ← -e t-s 解析学の平均値の定理の主張は,もしf(t) が微分可能な関数ならば、任意の実数 s ともに対し,stの間の値の実数0で次を満たすものが存在するというものである. f(t)-f(s) =f' (0) (t-s). もしweΩを固定し,f(t) = etx(w) を定義すると,この式は, etX(w)_esx(w)=(t-s) X (w)e (w)x(w) (1.9.1) となる.ただし,(ω) はωに依存する実数 (すなわち,tとsの間の値を取る確率変 数)である. (i) 優収束定理 (14.9) (191) 式を使って,次を示せ. lim EY = Elim Yn=E [XetX] . (1.9.2) n→∞ [n→∞ このことから,求める式 4'(t) [XetX ] が導かれる. (ii) 確率変数 X は正の値も負の値も取り得、全てのt∈Rに対し Eetx < かつ E [|X|etX] < ∞ であると仮定する。 再度 '(t) = E [XetX] を示せ(ヒント: (1.3.1) 式の記号を使って X = X + - X- とせよ . )

未解決 回答数: 1
1/15