ducs-personal/kdd2014-1

kdd 2014

kdd2014

数据说明

• donations.csv: 含有每个项目的donations信息。只有训练集有
• essays.csv: 教师提交的项目的文本。训练、测试都有
• projects.csv: 关于每个项目的信息。训练测试都有
• resources.csv: 含有每个项目需要的信息。训练测试都有
• outcomes.csv:训练集的结果
• sampleSubmission.csv:样例输出

"Exciting" Projects

Exciting projects 符合DC.org网站的各种需求。注意，“exciting” 意味着有商业贡献并且这并不意味着non-exciting project会对教师、学生、捐赠者产生影响(are not compelling to teachers/students/donors-这句我不会翻译)。为了成为exciting， 一个项目必须符合全部五个限制。下面是这五个限制，带有括号的是其在数据中的出现名称。

• 被成功完成(fully_funded)
• 有至少一个teacher-acquired donor(at_least_1_teacher_referred_donor)
• 有超过平均段的捐赠者留言 (great_chat)
• 至少有一个"green"捐赠 (at_least_1_green_donation)
• 有一个或者多个
  • 3个或更多非teacher-acquired捐赠者(three_or_more_non_teacher_referred_donors)
  • 一个非teacher-acquired教师给了超过$100 (one_non_teacher_referred_donor_giving_100_plus)
  • 项目受到了"thoughtful donor"的捐赠(donation_from_thoughtful_donor)

你将会在outcomes.csv中找到摘要信息，包括is_exciting变量的真值信息

Data fields

下面是提供的数据的一个简短解释。有一些很清楚的就不细说了

• outcomes.csv

  • is_exciting - 从商业的角度看，是否一个项目是exciting的
  • at_least_1_teacher_referred_donor - teacher referred = donor donated because teacher shared a link or publicized their page
  • fully_funded - 项目成功完成
  • at_\least_1_green_donation - 由一些大公司(Amazon)或者先进支付的捐赠
  • great_chat - 项目评论数超过平均
  • three_or_more_non_teacher_referred_donors - non-teacher referred is a donor that landed on the site by means other than a teacher referral link/page
  • one_non_teacher_referred_donor_giving_100_plus - see above
  • donation_from_thoughtful_donor - 就是一帮很刁的捐赠者
  • great_messages_proportion - great_chat 如何被算出。 如果大于现阶段的62%，great_chat=True
  • teacher_referred_count - teacher referred的数量 (see above)
  • non_teacher_referred_count - non-teacher referred的数量 (see above)

• projects.csv

  • projectid - 项目id
  • teacher_acctid - 创建项目的教师的id
  • schoolid - 教师工作的学校的id
  • school_ncesid - public National Center for Ed Statistics id
  • school_latitude - 学校维度
  • school_longitude - 学校经度
  • school_city - 学校城市
  • school_state
  • school_zip
  • school_metro - 地铁
  • school_district - 学校地区
  • school_county
  • school_charter - “特许”学校-网上查的
  • school_magnet - whether a public magnet school or not
  • school_year_round - whether a public year round school or not
  • school_nlns - whether a public nlns school or not
  • school_kipp - whether a public kipp school or not
  • school_charter_ready_promise - whether a public ready promise school or not
  • teacher_prefix - 教师性别
  • teacher_teach_for_america - Teach for America or not
  • teacher_ny_teaching_fellow - New York teaching fellow or not
  • primary_focus_subject - main subject for which project materials are intended
  • primary_focus_area - main subject area for which project materials are intended
  • secondary_focus_subject - secondary subject
  • secondary_focus_area - secondary subject area
  • resourc_type - main type of resources requested by a project
  • poverty_level - school's poverty level. highest: 65%+ free of reduced lunch high: 40-64% moderate: 10-39% low: 0-9%
  • grade_level - grade level for which project materials are intended
  • fulfillment_labor_materials - cost of fulfillment
  • total_price_excluding_optional_support - project cost excluding optional tip that donors give to DonorsChoose.org while funding a project
  • total_price_including_optional_support - see above
  • students_reached - number of students impacted by a project (if funded)
  • eligible_double_your_impact_match - project was eligible for a 50% off offer by a corporate partner (logo appears on a project, like Starbucks or Disney)
  • eligible_almost_home_match - project was eligible for a $100 boost offer by a corporate partner
  • date_posted - data a project went live on the site

下面还有很多，今天先拉到吧