GFQR 1026 Lecture 3: Big Data Analytics in Business (1) PDF

GFQR 1026: Big Data in “X” Lecture 3 : Big Data Analytics in Business (1) ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Page 11  Lecture 3: Outline Business Intelligence Types of B...

GFQR 1026: Big Data in “X” Lecture 3 : Big Data Analytics in Business (1) ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Page 11  Lecture 3: Outline Business Intelligence Types of Big Data Analytics used in Business Examples of Big Data Analytics in Business Technologies used in Business: Recommendation Systems & Cookies 2 2 Page ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Business Intelligence (BI) Businesses have been using business intelligence tools for many decades – Business intelligence consists of a broad category of applications and technologies for gathering, storing, analyzing, and providing access to data. – BI provides historical, current, and predictive views of business operations. Compile Business Intelligence with Big Data ! Page 3 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Revenue from Big Data and Business Analytics Source: https://www.statista.com/statistics/551501/worldwide-big-data-business-analytics-revenue/ Page 4 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data Analytics Big Data Analytics is used everywhere: Focus on Big Data in Business in this Lecture ! Page 5 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Types of Big Data Analytics in Business 1. Descriptive Analytics 2. Predictive Analytics How can we make it happen? 3. Prescriptive analytics Prescriptive Analytics (Decision) Predictive What Analytics happened? (Future) What will Descriptive happen? Analytics (Past) Page 6 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  1. Descriptive Analytics The most traditional of business analytics It accounts for the majority of all current business analytics Looks at past performance and understands that performance by mining historical data to look for the reasons behind past success or failure It uses data aggregation and data mining to answer: “What has happened?” – e.g. What is the average weekly sales amount? Page 7 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  2. Predictive Analytics Uses statistical models and forecasting techniques to understand the future and answer: “What could happen?” – e.g. Forecast the demand of smart phone sales in next year Provides companies with actionable insights based on data Provides estimates about the likelihood of a future outcome Page 8 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  3. Prescriptive Analytics Goes beyond predicting future outcomes by also suggesting actions to benefit from the predictions and showing the decision maker the implications of each decision option Not only anticipates what will happen and when it will happen, but also why it will happen Can suggest decision options on how to take advantage of a future opportunity or mitigate a future risk and illustrate the implication of each decision option Can continually and automatically process new data to improve prediction accuracy and provide better decision options e.g. Google self driving car Page 9 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Google self driving car Big data is used to operate Google self driving car https://www.youtube.com/watch?v=TsaES--OTzM Page 10 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Business Analytics Overview Page 11  Big Data in Aviation Industry Examples of Big Data: – Departure time and Arrival time of Flights – Activity on social media – In flight food preferences – The number of people you’re travelling with Usages – Aircraft Maintenance and Inventory Management – Fleet Management – Operations Management Benefits: – Pilot, Crew and Staff Management Smarter maintenance Improve safety Improve service Reduce cost Page 12 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Aviation Industry Case: United Airlines - Encourage loyalty – Use their “collect, detect, act” protocol to analyze over 150 variables in each customer profile. e.g. passengers with over 50 years old like to take domestic airlines. – These analyses measure everything from previous purchases to customer preferences in order to generate a tailor-made offer. – The “collect, detect, act” initiative has increased United’s year- to-year revenue by over 15%. Case: British Airways - Get to know the customer – Use an intelligent ‘Know Me’ feature to provide personalized search results to customers. e.g. Show early flight results to passengers who always take early flights. – Use in-depth data analysis to provide relevant and targeted offers to customers Page 13 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Aviation Industry Case: EasyJet - Deploy artificial intelligence – Invested in artificial intelligent that determines seat pricing automatically, depending on demand – Analyze historical data to predict demand patterns up to a year in advance. – For future decision-making about new routes, schedules etc. Case: Southwest Airlines - In-flight intelligence – Teamed up with NASA to continually improve airline safety – Southwest and NASA have created an automated system – To generate enormous amount of data to flag anomalies and prevent accidents Page 14 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Aviation Industry Case: Cathay Pacific – Use big data and IoT in various areas such as flight planning, fuel optimization and preventive maintenance – Enable it to stay competitive and enhance passenger experience – All aircrafts are equipped with flight monitors, the data captured is for safety analysis and corporate training – Working with aircraft engine provider Rolls Royce to create an analysis on how to optimize the operations of jet engines – Enhance customer experience by combining operational with sales and marketing data to offer personalized travel offerings to customers Page 15 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Retail Industry “The internet has entered the most critical 30 years in terms of application... Many industries will be redefined in the new era powered by cloud computing, Big Data and the internet of things,” Jack Ma Yun Case: Alibaba – One of the top winners of Big Data users in Asia – Online transaction of US$ 84.5 billion (HK$ 659 billion) on its annual Singles’ Day sale in 2021. – Use Big Data analytics to enable Alibaba and its sellers at TMall and Taobao to cross-sell and up-sell – To generate more and more transactions & revenues – Big Data includes: shopping habits, payment and credit history, demographics, search preferences, social networks, personal interests – Use real-time online data to predict what consumers want (recommendation system) Page 16 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Recommendation Systems The input to a recommendation system is the feedback of likes and dislikes, and the output is recommended items based on the feedback Examples – Netflix/Youtube: Movie/video recommendation – Amazon.com: “Customers who bought this item also bought” section – Spotify: Music recommendations – Google: News recommendations Page 17 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Two approaches to build Recommendation Systems Page 18 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Content-based Filtering Build recommenders based on item attributes – e.g. Item attributes in movies: the genre, actor, director, producer and hero A user’s taste identifies the values and weights for an attribute, which are provided as an input to the recommender system This technique is purely domain-specific & the same algorithm cannot be used to recommend other types of products e.g. Content-based filtering is to recommend movies in the western genre to a user who watches many cowboy movies Page 19 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Collaborative Filtering Recommend items based on similarity measures between items or users Items preferred by similar users will be recommended to another user – Like how you get recommendations from your friends This technique is not domain-specific and does not know anything about the items and the same algorithm can be used for any type of product (e.g. movies, music, books) Two Types: user-based and item-based Page 20 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  User-based Collaborative Filtering Based on the similarities between users The idea behind this algorithm: similar users share similar preferences e.g. User1 and User3 rated Movie1 and Movie4 with similar rating (4 & 5): indicates the taste of User1 and User3 is the same. As such, we can recommend Movie2 to User1, which is rated 5 by User3. Similarly, Movie3 can be recommended for User3 User1 User2 User3 Movie1 4 4 5 Movie2 4 5 Movie3 4 2 Movie4 4 5 Page 21 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Item-based Collaborative Filtering Based on similarities between items The idea behind the algorithm: a user will have a similar preference for similar items Works like this: For every I items that a user U has no preference for, – Compute the similarity between I and every other item that U has a preference for – Calculate a weighted average, where the weighted preference is the product of the similarity of item I with any other items that U has expressed a preference for, with the preference value for that item. – Adding this weighted preference for all items that U has a preference for gives the weighted sum, and dividing it by the number of such items gives the weighted average of preference value P. – The P value is the preference for item I for user U, and if this is above a particular threshold, we can recommend the item to U. Page 22 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Item-based Collaborative Filtering To build an item-based recommender, we need preference data and a notion of similarity between items Scalability of item-based collaborative filtering systems is much better than user-based filtering Item-based recommendation systems are most widely used Page 23 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Limitations of Recommendation System The cold-start problem – Collaborative filtering systems are based on the action of available data from similar users – If you are building a brand-new recommendation system, you would have no user data to start with. You can use content-based filtering first and then move on to the collaborative filtering approach Page 24 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Limitations of Recommendation System Scalability – As the number of users grow, the algorithms suffer scalability issues. – If you have 10 million customers and 100,000 movies, you would have to create a sparse matrix with one trillion elements Page 25 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Limitations of Recommendation System The lack of right data – Input data may not always be accurate because humans are not perfect at providing ratings – User behavior is more important than ratings – Item-based recommendations provide a better answer in this case Page 26 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Cookie They are small data files that are written to the user’s hard drive when the user browses a website or fills in a registration form Record the user's browsing activity Remember arbitrary pieces of information that the user previously entered into form Once stored, it can be re-accessed the next time the user visits the relevant websites, – e.g. to show previous orders or the last page visited, list out similar items that the customer may wish to purchase. Page 27 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Cookie Guidance for Data Users on the Collection and Use of Personal Data through the Internet Page 28 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Retail Industry Case: Chow Tai Fook Jewelry - identify shopping behavior – Massive amounts of data related to customers and products via different channels such as retail stores, e-commerce platforms, and membership program – Identify individual customers’ styles, preference, buying habits, predicting business performance and improving inventory turnover Case: Li & Fung in China - identify shopping behavior – Set up an experimental shopping mall in Shanghai called Explorium – Enhance shoppers’ experience by understanding their personal preferences better through technology – The shopping mall takes advantage of different state-of-the-art technologies and top-class big data analytics to track and identify shopper behaviors Page 29 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Big Data in Retail Industry Page 30 Big Data in Banking & Finance Industry Case: Hang Seng Bank – Accept online applications for personal loans without requiring borrowers to provide any proof of income or address – Use big data to analyze customer behavior and credit worthiness of the customers to decide if the bank should approve their personal loan applications Page 31 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Big Data in Banking & Finance Industry Case: Citibank – Over 200 million customer accounts globally – Analyzing data and targeting promotional spending – Scan transactional records to spot anomalies such as incorrect, unusual or fraudulent charges – Big Data-driven approach drives business growth and enhances the services provided to customers – Big Data analytics have been implemented successfully in customer retention and acquisition Page 32 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Big Data in Banking & Finance Industry Case: WeLend - operated by WeLab – Hong Kong’s first online lending platform, founded in 2013 – Make loans to customers without ever meeting them face to face – A loan approval takes just 1.7 seconds – Determine the creditworthiness of each user using its proprietary big data technology – In Hong Kong, the company relies primarily on a user’s credit history and interaction data – Collect a wider variety of information from its users, such as the phone model they use, installed apps, social networks and even how much time they spend talking on the phone – loan delinquency rate: just 1% in HK Page 33 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Insurance Industry Case: HSBC Insurance – Understanding client needs through big data analysis resulted in several product innovations – HSBC Term Protector (2016), and HSBC Cancer Term Protector (2017) – Understand the customers through digital analytics – Better meet customers’ needs Benefits and Usages – Smarter, cheaper products – Developing new products and risk management models, which would allow insurance companies to operate at a lower cost and understand risk – Predict illness, accidents and price their products accordingly Page 34 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Insurance Industry Case: MLC Life Insurance – Use wearable technology to record health activity and is designed to encourage Australians to lead healthier lives – More than 17% of all new policy holders have signed up to it and are reaping the rewards of better health and reduced life insurance premiums Opportunities – Product Innovation – Improved customer experience – Better engagement with customers – Risk understanding – room for improvement across the industry – Predictions – more effective in relation to likelihood of claims Page 35 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data in Insurance Industry Case: ZhongAn Online Casualty and Property Insurance – China’s first internet-only insurer – Big Data platform includes companies along the car value chain and provides a one-stop shop for buyers and owners, this helps ZhongAn boost its car insurance revenues – Can offer drivers better discounts based on their mileage and driving behavior – Insurers can also redesign policies for next year based on drivers’ behavior Page 36 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data used for Hospitality Industry Case: Marriott – Marketing, and hotel chain’s operations – Use unstructured and semi-structured data such as weather reports and local events schedules to forecast demand and determine a value for each individual room throughout the year – Set prices with optimum efficiency Case: Starwood Hotels and Resorts – 1,200 hotels around the world – Optimizing room pricing by analyzing data on local and worldwide economic factors, events and weather reports – Increase in its revenue-per-room of ~ 5% Page 37 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Big Data Applications in Business Sectors Airline: identify malfunction by using sensor, improve safety, smarter maintenance, reduce cost and improve customer service. Retail: use Big Data Analytics to optimize their business. Bank and Finance: use analytics to differentiate fraudulent interactions from legitimate business transactions, and suggest immediate actions, such as blocking irregular transactions, which stops fraud before it occurs and improves profitability. Insurance: for risk assessment, fraud detection, marketing, customer insights, customer experience and more. Hospitality: offer optimal price, data-driven decisions, launch personalized marketing campaigns. Page 38 ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Big Data around you ! 39 39 To be continued…. ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed. Page  Lecture 3: Summary Business Intelligence Types of Big Data Analytics used in Business Examples of Big Data Analytics in Business Technologies used in Business: Recommendation Systems & Cookies 40 40 Page ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  References 1. http://www.bdbanalytics.ir/media/1121/big-data-analytics_turning-big-data-into-big-money.pdf 2. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7100420 3. https://www.statista.com/statistics/551501/worldwide-big-data-business-analytics-revenue/ 4. https://www.edureka.co/blog/big-data-analytics 5. https://www.educba.com/predictive-analytics-vs-descriptive-analytics/ 6. https://www.youtube.com/watch?v=TsaES--OTzM 7. https://www.promptcloud.com/blog/five-interesting-use-cases-of-big-data-analytics-in-airline-industry/ 8. https://blog.datumize.com/5-relevant-examples-of-a-big-data-case-study-from-the-airline-industry 9. https://pxhere.com/en/photo/864346 10. https://www.scmp.com/tech/enterprises/article/2102193/alibaba-seeking-reshape-retail-industry-through-big-data-cloud 11. https://www.cw.com.hk/it-hk/hkcsview-big-data-action-hong-kong 12. https://www.alizila.com/at-alibaba-artificial-intelligence-is-changing-how-people-shop-online/ 13. Ankam, V. (2016). Big Data Analytics. Packt>, pp.214-216. 14. https://en.wikipedia.org/wiki/HTTP_cookie 15. https://www.pcpd.org.hk/english/publications/files/guidance_internet_e.pdf 16. http://cw.com.hk/opinion/big-data-action-hong-kong 17. https://www.cw.com.hk/it-hk/chow-tai-fook-charts-path-to-innovation 18. https://www.cw.com.hk/digital-transformation/fung-group-launches-explorium-2-0-shanghai 19. https://www.youtube.com/watch?v=3VWYKsaUezk 20. https://www.scmp.com/business/banking-finance/article/2152968/hang-seng-bank-relax-loans-rules-using-big-data-and-fintech 21. https://www.about.hsbc.com.hk/-/media/hong-kong/en/news-and-media/180315-hsbc-commercial-banking-uses-big-data-analytics-eng.pdf 22. https://www.bernardmarr.com/default.asp?contentID=1086 23. https://www.forbes.com/sites/bernardmarr/2016/09/09/big-data-in-banking-how-citibank-delivers-real-business-benefits-with-their-data-first- approach/#5c42964947e0 24. https://www.scmp.com/tech/start-ups/article/2119209/higher-rates-addresses-caps-online-lender-welend-reveals-how-it 25. http://www.startupshk.com/we-are-hk-founders-welab/ 26. https://www.edureka.co/blog/what-is-big-data/ 27. https://hongkongbusiness.hk/co-written-partner/sponsored-articles/hsbc-leading-insurance-era-innovation 28. https://www.mlc.com.au/content/dam/mlc/documents/pdf/media-centre/big-data-report.pdf 29. https://www.forbes.com/sites/bernardmarr/2019/05/27/this-health-insurance-company-tracks-customers-exercise-and-eating-habits-using-big-data-and- iot/#35716a476ef3 30. https://www.scmp.com/business/companies/article/2131948/big-data-and-ai-future-car-insurance-according-chinas-first 31. https://www.forbes.com/sites/bernardmarr/2016/01/26/how-big-data-and-analytics-changing-hotels-and-the-hospitality-industry/#539a06a71c22 32. https://www.edureka.co/blog/big-data-analytics/#Use%20Cases 33. Alibaba’s 2020 Singles Day Breaks Record, Attracts Luxury Brands | Practical Ecommerce 34. Alibaba (BABA) Singles’ Day Posts Record Sales - Bloomberg 41 41 Page ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.  Image Credits [Slide 2 & 41] https://www.kissclipart.com/instruct-clipart-clip-art-women-clip-art-zameoc/ [Slide 3] https://pixabay.com/illustrations/big-data-information-business-2426949/ [Slide 4] https://www.statista.com/statistics/551501/worldwide-big-data-business-analytics-revenue/ [Slide 5] https://www.edureka.co/blog/big-data-analytics/#Why%20Big%20Data%20Analytics? [Slide 9] https://www.forbes.com/sites/brookecrothers/2015/11/12/google-is-leader-in-revolutionary-self-driving-cars-says-ihs/#6fbb1cca39b7 [Slide 15] https://pxhere.com/en/photo/864346 [Slide 17] https://media.netflix.com/en/company-blog/2020-netflix-recommended-televisions [Slide 17] https://www.slideshare.net/mounialalmas/recommending-and-searching-research-spotify [Slide 18] http://datameetsmedia.com/an-overview-of-recommendation-systems/ [Slide 23] https://medium.com/@cfpinela/recommender-systems-user-based-and-item-based-collaborative-filtering-5d5f375a127f [Slide 24] https://towardsdatascience.com/build-your-own-recommender-system-within-5-minutes-30dd40388fbf [Slide 25] https://ebaytech.berlin/deep-learning-for-recommender-systems-48c786a20e1a [Slide 27] https://www.comforte.com/enterprise-data-protection/big-data-security/ [Slide 33] https://pixabay.com/photos/bank-money-finance-shares-save-2907728/ [Slide 35] https://www.researchgate.net/figure/Different-types-of-wearable-technology_fig5_322261039 [Slide 37] https://www.the-digital-insurer.com/dia/zhong-an-chinas-first-complete-online-insurance-company/ [Slide 42] https://www.edureka.co/blog/big-data-applications-revolutionizing-various-domains/ [Slide 13, 14, 16, 29, 31, 32, 34, 35, 36, 38 ] Wikipedia 42 42 Page ©COMP, HKBU 2020. All Rights Reserved. This content is copyright protected and shall not be shared, uploaded or distributed.