大数据Hadoop技术在银行的七个应用实例

引读：如今，hadoop几乎存在于各个方面，其通过利用大[color=rgb(105, 105, 105) !important]数据来分析信息和增加竞争力。许多金融机构和公司已经开始使用Hadoop成功地解决问题，即便他们本没有计划这样做。因为如果他们不这样做，就会面临市场份额损失的巨大风险。以下是一些特别有趣和重要的大数据和Hadoop用例。

诈骗侦测（Fraud detection）：诈骗是金融犯罪和数据泄露中成本最大的挑战之一。Hadoop分析可以帮助金融机构检测、预防和减小来自内部和外部的诈骗行为，同时降低相关成本。销售、授权、交易以及其他的[color=rgb(105, 105, 105) !important]数据分析点能够帮助银行识别和减少诈骗。例如，大数据技术通过提取异常行为模式，能够提醒银行信用卡或借记卡的丢失或被盗。从而给银行提供时间暂时冻结异常账户，直到联系到账户持有人为止。

风险管理（Risk management）：任何一家金融公司都需要准确地评估风险，而大数据解决方案就使他们能够有效地评估信贷风险。银行分析交易数据，基于模拟市场行为、评估[color=rgb(105, 105, 105) !important]用户和潜在用户，来确定风险和泄露。Hadoop的解决方案对风险和后果具有全面而准确的考虑，使企业能做出最好、最明智的决策。

客服中心效率优化（Contact center efficiencyoptimization）：确保用户满意无疑是最重要的。涉及到金融业，大数据允许银行提前预测用户需求用以快速地解决问题。客服中心的数据分析提供了媒介，及时、简洁的洞察力，能够快速满足用户的需求，从而确保了效率成本甚至提高了交叉销售的成功率。

客户分类优化产品（Customersegmentation for optimized offers）：大数据提供了一种方法从粒度级别来理解客户的需求，以至于银行和金融机构能够更有效地提供有针对性的优惠。转而，这些更加个性化的产品带来更高的接受度，提高客户的满意度，制造更高的利润和更好的客户保留。来自于社交媒体和交易的顾客详细信息则可以用来降低用户的采购成本以及周转率。

客户流失分析（Customer churn analysis）：大家都知道开发新客户比留住老客户的成本要高，大数据和Hadoop技术可以通过导致客户放弃的行为分析和识别模式来帮助金融公司来留住他们的客户。什么时候客户会最可能因为竞争对手而离开？什么原因？导致客户不满意的因素是什么？公司失败在哪里？这些决定如何避免客户放弃的信息都是无价的。为了迎合客户需求，使客户利益最大化，学习用正确的步骤来执行对金融公司公司来说势在必行。

情感分析（Sentiment analysis）：Hadoop和先进的分析工具有助于分析社会媒体来达到监控企业用户的情感，品牌或产品的目的。如果一家银行参加竞选，大数据工具可以通过名称，和标签报告以及竞选活动名称或平台报告来监控社会媒体。细节分析是富有洞察力的，银行可以基于这些根据时间，目标和人口特征的见解来准确地做出决策。

客户体验分析（Customer experience analytics）：作为面向客户的企业，金融机构需要利用到存于各种业务线筒仓的客户数据。这些包括投资组合管理，客户关系管理，贷款系统，呼叫中心等等。大数据可以提供更好的洞察和理解，帮助公司迎合客户需求以及前景需求。这些都可以帮助企业优化提高利润，并维护长期的客户关系。

底线是所有的企业，尤其是金融公司，需要使用大数据和Hadoop技术充分发挥他们的潜力，特别是对于每天交易所积聚的海量数据。为了保持竞争力，维持现有客户并吸引新客户，金融公司应该从今开始计划使用大数据技术，否则会因为竞争对手对这些技术的使用而失去更多的客户。那并不意味着要使用每一个可行的方式— 而只是运用对每个机构最好的可行方式。

大数据和Hadoop技术非常强大，可帮助金融机构在市场上保持领先。运用了这些技术就能看到他们传输的结果。

英语原文：

Hadoop is present in nearly every vertical today that isleveraging big data in order to analyze information and gain competitiveadvantages. Many financial organizations firms are already using Hadoopsolutions successfully and the ones who are not have plans to do so. If theydon’t, they risk enormous market share loss. Followingare a few of the most intriguing and essential big data and Hadoopuse cases.

Fraud detection:Fraud, financial crimes and data breaches are some of the most costlychallenges in the industry. Hadoop analytics help financial organizationsdetect, prevent and eliminate internal and external fraud, as well as reducethe associated costs. Analyzing points of sale, authorizations andtransactions, and other data points help banks identify and mitigate fraud. Forexample, big data technology can alert the bank that a credit or debit card hasbeen lost or stolen by picking up on unusual behavior patterns. This then givesthe bank time to put a temporary hold on the card while contacting its accountowner.

Risk management:Every financial firm needs to assess risk accurately, and big data solutionsenable them to to do so by effectively evaluating credit exposures. Banksanalyze transactional data to determine risk and exposure based on simulatedmarket behavior, scoring customers and potential clients. Hadoop solutions allowfor a complete and accurate view of risk and impact, enabling firms to make thebest, most informed decisions.

Contact center efficiencyoptimization: Ensuring customers are satisfied is of utmostimportance when it comes to finances, and big data can help resolve problemsquickly by allowing banks to anticipate customer needs ahead of time. Analyzingdata within the contact center provides agents with timely and concise insightthat satisfies customers quickly and efficiently, ensuring cost effectivenessand even improving cross-sales success rates.

Customer segmentation foroptimized offers: Big data provides a way to understand customers’ needs at a granular level so that banks and financial organizationscan deliver targeted offers more effectively. In turn, these more personalizedoffers result in higher acceptance rates, increased customer satisfaction,higher profitability and greater retention. Detailed information aboutcustomers derived from social media and transactions can be utilized to reducecustomer acquisition costs as well as turnover.

Customer churnanalysis: Everybody knows that it’s cheaper to keep acustomer than it is to go out and find a new one. Big data and Hadooptechnologies can help financial firms keep retain more of their customers byanalyzing behavior and identifying patterns that lead to customer abandonment.When are customers most likely to leave for the competition, and why? Whatcauses customer dissatisfaction? Where did the firm fail? This information fordetermining how to avoid customer abandonment is priceless. It’s imperative for financial firms to learn the right steps toimplement in order to meet customer needs and save their most profitablecustomers.

Sentimentanalysis: Hadoop and advanced analytics tools help analyze social media inorder to monitor user sentiment of a firm, brand or product. If a bank isrunning a campaign, big data tools can monitor social media by name and reporton it by hashtag, campaign name or platform. Analytics on the fine-graineddetails are insightful, and the bank could then make decisions more accuratelybased on these insights in terms of timing, targeting and demographics.

Customerexperience analytics: As consumer-facing enterprises,financial institutions need to take advantage of the customer data that residesin all of the silos across various lines of business. These include portfoliomanagement, customer relationship management, loan systems, contact center,etc. Big data can provide better insight and understanding, allowing firms tomatch offers to a customer or prospect’s needs. This thenhelps the firm to optimize and improve profitable and long-term customerrelationships.

The bottom line is that allenterprises, especially financial firms, need to use big data and Hadooptechnologies to their fullest potential now, particularly with the overwhelmingamount of data and transactions amassed on a daily basis. In order to remaincompetitive and maintain current customers while attracting new ones, financialfirms should start planning to utilize big data technologies today or risklosing more customers to competitors utilizing these tools. That doesn’t necessarily mean in every way possible –it just means in the best way possible for each organization.

Big data and Hadoop technologies arepowerful and help financial organizations stay ahead in the market. Set them inmotion and watch them deliver results.