今年の総引用数予測アプリの使い方

Google Scholarなどで各年の論文の引用数を調べて、「一昨年の総引用数」「昨年の総引用数」「今年の今日までの引用数」を入力してください。その後、「引用数予測値の計算」ボタンを押すと、今年の年末までの総引用数を予測できます。自分や知り合いの研究者の引用数を予測して楽しんでみてください。

---

一昨年の総引用数:

昨年の総引用数:

今日までの今年の引用数:

引用数予測値の計算

from datetime import datetime import math def clicked(): tot_cite_2nd = float(Element("tot_cite_2ly").element.value) tot_cite_1st = float(Element("tot_cite_1ly").element.value) cur_cite = float(Element("cur_cite").element.value) y1_log = math.log(tot_cite_1st + 1) y2_log = math.log(tot_cite_2nd + 1) y_log_exp = 2.0 * y1_log - y2_log cite_exp_1 = math.exp(y_log_exp)-1 today = datetime.today() end_of_year = datetime(today.year, 12, 31) start_of_year = datetime(today.year, 1, 1) days_left = (end_of_year - today).days days_total = (end_of_year - start_of_year).days + 1 rate = (days_total - days_left) / days_total if rate == 0: rate = 1 / days_total cite_exp_2 = cur_cite / rate f1 = tot_cite_2nd f2 = tot_cite_2nd + tot_cite_1st f2h = tot_cite_2nd + tot_cite_1st + cur_cite h = rate c = ((f2h/(2+h))-f1)/(1+h)-(f2/2-f1) c = c/(((3+4*h+h**2)/(1+h))-3) b = f2/2 - f1 - 3*c a = f1 - b - c t = 3 cite_exp_3 = max(a*t + b*t**2 + c*t**3 -f2, 0) cite_exp_mid = cite_exp_1 + cite_exp_2 + cite_exp_3 cite_exp_mid = cite_exp_mid - max(cite_exp_1,cite_exp_2,cite_exp_3) cite_exp_mid = cite_exp_mid - min(cite_exp_1,cite_exp_2,cite_exp_3) if(rate < 0.25): cite_exp = (rate/0.25)*cite_exp_mid + (1-(rate/0.25))*cite_exp_1 elif(rate < 0.75): cite_exp = (1-((rate-0.25)/0.5))*cite_exp_mid + ((rate-0.25)/0.5)*(cite_exp_2+cite_exp_3)/2 else: cite_exp = (cite_exp_2+cite_exp_3)/2 Element("output").write(f"今年の総引用数の予測値: {cite_exp:.2f}")