Facebook Recruiting IV: Human or Robot? Facebook招聘IV：人还是机器人？

有没有想过在Facebook工作是什么感觉?Facebook和Kaggle将在2015年举办一场工程竞赛。一路走到排行榜的顶端，获得一个软件工程师的面试机会，研究世界级的机器学习问题。

在这个比赛中，您将为一个在线拍卖网站寻找机器人。与软件控制的竞标者相比，该网站上的人工竞标者无法赢得竞标变得越来越沮丧。结果，网站核心客户群的使用量直线下降。

为了重建客户满意度，网站所有者需要从他们的拍卖中消除电脑产生的投标。他们试图用行为数据(包括短期内的投标频率)建立一个模型来识别这些投标，但这一尝试被证明是不够的。

该竞赛的目标是识别由“机器人”放置的在线拍卖出价，帮助网站所有者轻松标记这些用户，以便从他们的网站上删除，以防止不公平的拍卖活动。

这场比赛的数据来自一个在线平台，而不是Facebook。

请注意:你必须以个人身份参加招聘比赛。您只能使用所提供的数据进行预测。

Ever wondered what it would be like to work at Facebook? In 2015, Facebook and Kaggle hosted an engineering competition. Climb to the top of the leaderboard to earn a software engineering interview opportunity and tackle world-class machine learning problems.

In this competition, you will detect bots for an online auction website. Human bidders on the platform have grown increasingly frustrated when they fail to win bids against software-controlled competitors. As a result, usage among the website’s core customer base dropped sharply.

To rebuild customer satisfaction, the website operator needs to eliminate computer-generated bids from their auctions. They previously attempted to build a model to identify these bids using behavioral data (including short-term bidding frequency), but this approach proved insufficient.

The goal of this competition is to identify online auction bids placed by "bots", helping the website operator easily flag these users for removal from the platform to prevent unfair auction activities.

The data for this competition is sourced from an online platform, not Facebook.

Please note: You must participate in the recruiting competition as an individual. You may only use the provided data to generate your predictions.

搜集汇总

背景与挑战

背景概述

该数据集源自2015年Facebook与Kaggle联合举办的工程竞赛，目标是识别在线拍卖中由机器人发出的出价，以帮助网站消除不公平竞拍行为。数据集包含竞标者信息（如ID、付款账户）和出价记录（共760万条），涵盖拍卖、设备、时间等字段，用于训练和测试分类模型。