Predictive Analytics and Fantasy Football

by lowes1 on September 6, 2010

EWeek is reporting that Fantasy Football Secret Weapon is: IBM Predictive Analytics

IBM’s predictive analytics has proven to be not only a way for enterprises to make better business decisions, but to give fantasy football players a leg up in their leagues.

According to the report, If you are one of the 21 million people who will sign up for fantasy football this year, IBM’s predictive analytics could be the secret weapon you need to win in your league.

According to Hetal Thaker, a product manager at IBM and long-suffering fan of the Detroit Lions, using IBM’s predictive analytics technology has boosted her game, leading her to win three times in the last five years. Yes, she’s a girl and she’s been using IBM’s technology and whipping the boys at their own game. Short of conjuring up Barry Sanders to come back and suit up for her beloved Lions, Thaker has to rely on IBM predictive analytics to see her team win consistently.

In an interview on IBM’s new Predictive Perspectives blog, Thaker said, “Before using predictive analytics, I won once in 11 years. By applying this technology, I’ve won three times in the last five years. Nothing like dominating the league and holding bragging rights.”

And the girls are bringing it even more than the guys, Thaker said.

“Girls, at least the ones in our league, tend to be more analytical,” Thaker said. “They look at the data and read news stories. Some even make flash cards for the draft. Men tend to read a few things here and there, but think they know everything and end up drafting with their gut. That’s not to say there isn’t some degree of gut with the girls … also known as a football player’s ‘cuteness factor.’”

Yet, when Thaker realized she was always going back to historical data, such as evaluating performance in previous seasons or injury history, to determine her picks, she quickly realized there are products and technologies she could use to not only automate the analysis all of this data, but gain a better probability on—and predict—who would succeed.

Moreover, IBM’s text analytics tool allows Thaker to quickly analyze this data and categorize it. For example, she is able to get the latest news reports on all the players right before her draft and quickly auto-categorize the information into negative—critical injuries or suspensions—to positive—good preseason performance or good supporting cast. With these more accurate player profiles, she can now enter the draft with more confidence in who she thinks will yield the greatest output.

Thaker said textual data is the most important to her for analyzing players. According to Thaker:

“It’s really the textual data, such as injuries or past behavioral information, which can greatly improve my models and make them more accurate. Just like in the business world, combining structured information—transactional or demographic data—with attitudinal data—opinions, likes and dislikes—equates to having greater knowledge of a customer so an organization can create better marketing campaigns and targets. Textual information is very rich and provides deeper insight, but is also the hardest information to analyze, so it’s often excluded. I had the good fortune of having a robust text analytics tool that allowed me to quickly analyze this data and categorize it.”

With 21 million people taking part in fantasy football, it is big business with the likes of ESPN, the National Football League (NFL), CBS Sports and Yahoo, to name a few brand names sponsoring fantasy football platforms and monetizing the opportunity. It also is big entertainment and helps drive the popularity of pro football and the NFL brand.

Yet, while this is a fun look at what is possible with the power of data analysis, businesses should sit up and take notice, Thaker said. Any industry can benefit from using predictive analytics to improve marketing campaigns, identify fraud, reduce crime or find cures for diseases.

Thaker said businesses should take note and tap into the data analysis wave. Of the power of data analysis, Thaker said: “The uses are endless, even fantasy football. But, in all reality, if I’m able to put together a winning strategy for football, then why aren’t more organizations—or my league members—taking advantage of this? We all want to win, so why not create the best possible competitive advantage and ensure you’re making the best and smartest decisions.”

{ 0 comments }

Predictive Analytics and Crime

by lowes1 on August 14, 2010

News reports reflect that that Predictive Analytics assist with Fighting Crime in Chicago, Il.

Chicago Police Supt. Jody Weis talked Sunday about a new technique the department is using to fight crime.

It’s called Predictive Analytics. Data on crimes is combined with human intelligence to zero-in on places where crime is most likely to occur. That helps the department decide where to put its resources.

Predictive Analysis was launched at the end of April, and Weis says the department is already seeing positive results.

“If you can predict crime, even on a 50-percent basis, that goes a long way as you’re deploying your resources. And it can put you one step ahead of the criminals, which is what we’re trying to do,” Weis said.

The department has formed a predictive analysis group. CPD says it is always working with the Illinois Institute of Technology to develop models for predictive policing.

“This is a paradigm shift in how the department uses information,” said Chicago Police Supt. Jody Weis. “It will make for smarter policing.”

Launched in April, the project is funded by a $200,000 grant from the National Institute of Justice.

Weis explained how the information pinpoints the time of day and location of violent crimes. In the past, gang data were compiled on a yearly basis, Weis said.

Now any police interaction with a gang member will be fed into the system for a “real time, ongoing” account of where gangs operate.

{ 0 comments }

Law Firms Practice Predictive Analytics

by lowes1 on August 9, 2010

Law Firms Practice Predictive Analytics Says Lawyer Weekly

According to Canada’s Lawyer Weekly, when it comes to hiring new associates, law firms typically follow a very structured process. Often this involves hiring a large pool of summer students, narrowing down that pool when selecting articling students and then further tightening the assessment criteria when identifying new associates. This tiered approach is designed to help firms identify the top performing students most likely to contribute to the firm’s long-term productivity.

The article published this week states that in practice, however, this isn’t always the case. Students who perform well during summer internships and articling, may not contribute as much to the firm’s financial growth as expected. It can be difficult to decide who will be a top performer in your law firm based solely on grades and preliminary interviews. There are many qualities that come into play including, having the ability to attract the right clients and provide a consistent level of customer service. Lawyers need to be able to achieve their billable hour targets as well as adapt to the firm’s culture.

Why it’s so hard to hire right
When you consider how rapidly the workplace is shifting, it’s no surprise that firms can have difficulty hiring the right people up front. On the one hand, today’s young lawyers continue to expect quicker advancement opportunities and greater mobility than ever before. On the other hand, as we know, Canada’s average workforce population age is increasing.

Due to these trends, law firms increasingly need the ability to manage a multi-generational workforce. At the same time, they need strategies for identifying and retaining those lawyers best suited to lead the firm into the future. Unfortunately, some law firms lack formal systems for identifying emerging leaders. Current hiring practices rarely take into account the need to hire specific talent to fill expected turnover, support new service offerings or meet fluctuating demand within each practice group.

In an effort to overcome some of these process gaps, law firms have adopted a range of strategies. Many firms try to improve their associates’ chances for success through formal mentoring, training and development programs. They aim to retain high-performing talent through compensation and rewards, job flexibility and other perks.

Each of these approaches form a critical pillar in a law firm’s talent management arsenal. Alone this is not enough. To manage growth, maintain profitability and mitigate the risks of ongoing volatility, law firms should also adopt more sophisticated approaches for predicting their current and future talent requirements. Here are four methods to consider:

1. Workforce analytics
Workforce analytics use human resources data to help firms make informed decisions around hiring, promotion and pay. Use metrics?—?such as performance reviews, salary levels, billable hours, revenue by practice group and even the average age of practitioners?—?to assess the effectiveness of your firm’s hiring strategies and compensation programs. The data may offer answers to critical questions such as the average level of your turnover over the past few years, the best hiring sources and the highest-performing departments. Armed with this knowledge, you can begin to optimize your hiring strategies to attract specific people who possess the characteristics that traditionally contribute to your firm’s success.

Before workforce analytics can pay dividends, you need to ensure the quality of your firm’s data. Using incomplete, inaccurate or out-of-date data will hamper your abilities to clearly understand the factors that contribute to your workforce strengths and weaknesses.

2. Resource forecasting
Resource forecasting allows you to predict the future levels of demand for—and supply of—talent within each of your practice groups, at a local, national or even global level. By analyzing current workforce and economic trends, as well as organizational factors, you can often determine what skills your firm will need in the future. You can also assess if you have the internal resources to fill those needs, or if you will need to attract new external hires. By gaining a firm understanding of the number of people you will need in various practices, you will be better placed to allocate the right number of lawyers to each practice group.

3. Skill-based forecasting
An even more advanced approach to talent planning is the adoption of skill-based forecasting for key talent segments. With skill-based forecasting your law firm can start to identify those lawyers who possess the required skills to support the firm’s long-term strategy. By identifying people with the right skills, you can further foster their leadership development by assigning them to critical roles or placing them in a variety of cross-functional roles. In addition to plugging talent gaps, this can help maximize the firm’s growth and profitability over time.

4. Predictive modeling
A particularly sophisticated approach to talent planning involves statistically analyzing a firm’s data to determine the likelihood of, and reasons for, future events such as attrition. Using predictive analytics and retention modeling, firms can often identify those lawyers who are most at risk of leaving and which practice groups or geographic areas may be most subject to these talent shortages. By creating a profile of associates and partners most likely to stay or go, your firm can begin to predict potential vacancies and leadership gaps — and take steps to mitigate these risks in advance. Ultimately, understanding the causes of turnover allows you to reduce attrition, along with the financial and productivity costs associated with it.

The power of foresight
Although workforce analytics, forecasting and predictive modeling do not offer a 20/20 view into the future, they can provide law firms with a glimpse of the factors that may affect talent planning over time. This insight can position your firm to do more than determine the optimal composition of its future workforce needs. It can also enhance your decision-making when it comes to your most important asset: your people.

Sara Arnstein is a manager and Richard Lee is a partner in Deloitte’s Human Capital practice, based in Toronto. Deloitte’s Human Capital practice helps organizations develop and implement effective talent management strategies.

{ 0 comments }

Google, CIA fund predictive analytics firm

by lowes1 on August 4, 2010

Google, CIA fund predictive analytics firm says CNET

CNET is reporting Google Ventures and In-Q-Tel, the investment arm of the CIA, have provided funding to a company that monitors all the noise on the Web looking for connections between people, groups, and events, according to Wired.

The company, Recorded Future, offers a Temporal Analytics Engine for predictive analysis, allowing people to “visualize the future, past, or present.”

In addition, In-Q-Tel and Google Ventures both have seats on the board Recorded Future and have been “very helpful,” providing advice to the Cambridge, Mass.-based start-up, Chief Executive Christopher Ahlberg, an ex-Swedish Army ranger, told Wired in an article this week.

The amount of the investments is undisclosed, but it was less than $10 million each and was given in 2009, the article says.

This may be the first time Google and the CIA have funded the same firm, but it’s not the first time they’ve worked together. Google has sold servers to intelligence agencies and reportedly sought aid from the NSA after it was targeted in attacks it said originated in China. In-Q-Tel also had provided backing to Keyhole before Google acquired the mapping company to use in its technology in Google Earth.

{ 0 comments }

Guardian Analytics Launches Industry Roundtable Series to Drive Information Sharing Among Financial Institutions, Law Enforcement and Industry Experts

Discussions at inaugural event highlighted key trends and prevention strategies for cyber attacks against online banking

LOS ALTOS, Calif., July 28 /PRNewswire/ — Guardian Analytics, the innovator in predictive analytics-based fraud prevention software, recently hosted an exclusive industry event in San Francisco, bringing more than twenty executives from local banks and credit unions together with security, cybercrime, online banking and payments experts. Special guest speakers included the FBI, Jerry Silva, former bank executive and Tower Group analyst, and Joseph Menn, Financial Times writer and author of Fatal System Error. The educational and information sharing session focused on the challenges financial institutions face in growing the online channel and protecting it from rapidly advancing cybercrime techniques.

“Cyber criminals have developed the infrastructure to share information with each other, making them a powerful, fast-moving enemy,” said Terry Austin, CEO of Guardian Analytics. “Our goal is to help financial institutions better protect their customers from fraud that stems from cyber attacks with our fraud prevention offerings. We are committed to creating information and best practice-sharing forums such as our Industry Roundtable to strengthen the knowledge and effectiveness of our network of leading edge financial institutions.”

Roundtable participants shared trends in threats and fraud attacks against the online and mobile channels and discussed trends in the online and mobile channels, the need to balance security with customer convenience, and the use of customer data as they key to defending against new and emerging cyber attacks. They also discussed how they use FraudMAP® for Retail Banking and FraudMAP for Business Banking to proactively detect account takeover, even when complex Man in the Browser attacks are in play.

Trends in Online Business Banking Threats

A key takeaway from the event was that the institutions at the Roundtable continue to see fraudsters aggressively targeting their business accounts. Other observations include:

* Fraudsters are accessing online accounts through a trusted IP address with a significantly higher frequency. This change highlights the need to assume end user machines are compromised and to use online activity data to proactively detect account takeover
* Malware on users’ machines that facilitates Man in the Browser attacks (e.g. Zeus) is increasing. In some cases, multiple variations were found on the same machine
* Fraudsters are defeating dual controls regularly by fraudsters adding new users or changing approval limits on existing users
* Fraud alerts are frequently re-routed or deleted by fraudsters, keeping businesses in the dark that new users are being created or money is being moved
* Money mules continue to be a popular method to rapidly move large sums of money in smaller batches

Trends in Online Retail Banking Threats

Fraudsters are also advancing their methods to gain access to and steal money from retail accounts. Common threats and attacks discussed include:

* Using information gained from an online account compromise (personal data, signatures, account numbers) to perform offline wire transfers
* Using the forgotten password feature in online banking to compromise accounts
* Fraudulently opening online access for legitimate accounts that had not previously been using online banking, followed by rapid movement of large sums of money

Common Threads across Retail and Business Banking

While the attack methods are different between retail and business banking, some common themes emerged:

* Almost all attacks included some amount of account reconnaissance, signifying that banks typically have time in between account compromise and attempted money transfer to proactively stop fraudsters in their tracks
* Fraudsters are taking the time to learn each institution’s online banking platforms as well as security and payments processes and procedures in order to expedite their efforts.
* Fraudsters often prey on victims just before holiday weekends, hoping bank holidays will provide more time for them to move money unnoticed

“The information shared at the Industry Roundtable demonstrates that conventional wisdom is not enough to stop fraudsters,” continued Austin. “Common approaches like end user education, endpoint protection, dual controls, IP and geolocation-specific rules, and advanced authentication are controls that are not enough to prevent losses. Despite seeing such a complex and variable array of attacks, our customers were able to use the behavioral analytics in FraudMAP to protect themselves and their customers from losses.”

About Guardian Analytics

Headquartered in Los Altos, Calif., Guardian Analytics is focused on the prevention of online account fraud. The company’s real-time risk management approach to fraud detection, forensics and risk monitoring is built on strong analytics and predictive models of individual behavior. Leading financial services institutions rely on Guardian Analytics to protect individual account assets and the integrity of their online channels. Founded in 2005, Guardian Analytics is privately held with venture funding from Foundation Capital and Sutter Hill Ventures. For more information, please visit www.guardiananalytics.com.

{ 0 comments }

Memphis Cuts Crime With Predictive Analytics Reports Reflect

Recent press reports from Information Week reflects that Tennessee city attributes 31% drop in crime rate to knowing when and where to put cops on the street.

What started as an experimental predictive crime-prevention initiative in 2005 is now a proven program that’s about to get a big boost with new data and new forms of analysis.

IBM announced on Wednesday that the Memphis Police Department (MPD) is enhancing a predictive-analytics-based “Blue CRUSH” (Criminal Reduction Utilizing Statistical History) program by adding public call data to the information it analyzes to thwart crime before it occurs.

Mark Armstrong, president of WhamTech, discusses how the company helps customers such as the government integrate their disparate data sources. Microsoft Windows Server Group Product Manager, Manlio Vecchiet answers questions about Hyper V, network access protection and the adoption to date of Windows Server 2008. We caught up with CEI’s CEO D. Raja to talk about the state of the custom development market, what CEI’s role in it is and also about the entrepreunerial nature of the Pittsburgh PA market.
We caught up with CEI’s CEO D. Raja to talk about the state of the custom development market, what CEI’s role in it is and also about the entrepreunerial nature of the Pittsburgh PA market.
Launched five years ago by the University of Memphis Department of Criminology and Criminal Justice, the Blue CRUSH program started with analysis of historical data on gun crimes committed in North Memphis. Using predictive analytics software from SPSS (which was acquired last year by IBM), the University modeled patterns of crime and shared the results with area law enforcement agencies.

“Of course we knew the areas that had a lot of gun-related crime, but [Blue CRUSH] analyses helped us see patterns of exactly when and where the incidents were occurring,” said John Williams, Crime Analyst Unit Manager at MPD. With better insight, the department was able to send patrols to the right places at the right times, increasing arrest while also curbing incidents, Williams said.

Some consumers remain wary to conduct mobile transactions but perception, reality aren’t in sync
The State of Mobile Security

Early successes with Blue CRUSH led MPD to partner with the University as well as county and federal law enforcement agencies in 2006, pooling data and sharing insights on crime patterns. Analyses gradually expanded citywide, and the system now works in tandem with an MPD Real Time Crime Center (RTCC) monitoring and analysis hub opened in 2008.

Each week, the last four weeks’ worth of crime data is joined to spatial geographic information system data. Incidents are then plotted on digital map of the city within the RTCC that is updated and monitored around the clock. Blue CRUSH analyses of 28-day and seven-day statistics pinpoint crime hot spots, with details down to the day of the week and times of the day that are most active.

Blue CRUSH is credited as a primary driver of a 31% reduction in serious crime in Memphis since 2006, but Williams said MPD hopes to do better. Predictive analyses are currently based strictly on crime reports submitted by police officers. Not included are the more than two million emergency and non-emergency calls received from the public each year. MPD is now adding these call records to the Blue CRUSH analysis.

“There are some parts of town where we get reports of shots fired, gang members hanging out, or drug activity, but when we get there, the subjects are gone and there are no arrests,” Williams explained.

Call-volume and call-pattern analyses will not only uncover hot spots that don’t show up in arrest reports, it will help the department predict where it can put resources and change its approach to be more effective.

“If we can deploy some of our unmarked cars and undercover officers in these areas, we’re going to make arrests,” Williams said.

InformationWeek has published an in-depth report on energy-efficient government data centers. Download the report here (registration required).

{ 0 comments }

Predictive Analytics Gains More Market Acceptance

In this three minute interview, Phil Ice, Director Research and Development Public talks about how Predictive Analytics is gaining more acceptance at the university level.

 
icon for podpress  Flash Video: Play Now | Play in Popup | Download

{ 0 comments }

 
icon for podpress  Flash Video: Play Now | Play in Popup | Download

To benefit from Predictive Analytics, there are potential pitfalls to avoid along the way. In this interview with Vijay Desai, Ph.D., Principal Scientist, SAS, Dr. Desai shares his ideas and solutions as they compare across multiple industry segments.

Transcript:

Bill Cullifer: I am on the phone with Dr. Vijay Desai, Principal Scientist for SAS, S-A-S. Dr. Desai, good afternoon and thanks for agreeing to the call.

Vijay Desai: Thank you for having me.

Bill Cullifer: You bet. You recently participated in a panel discussion entitled Cross Industry Challenges and Solutions regarding predictive analytics and the theme included hurdles to jump and barriers to avoid when considering benefiting from predictive analytics. Can you summarize that discussion and what you walked away with?

Vijay Desai: Sure. Some of the things that we discussed were #1 analytics is becoming more and more useful in different industries and so as we go from industries where analytics has been used for a long period of time, in new industries there are some challenges. Some of these challenges include first convincing the users that predictive analytics can be of help, that predictive analytics is not going to take away their jobs, but by increasing their productivity it’s going to make them more valuable and help them generate more value and then even the data is becoming more and more available in different industries, you do not have what I would call targets. So, for example, in the case of fraud, you… target would mean identify the accounts that are good or fraudulent accounts. So, when you do not have targets available, it makes your job more difficult. So, in credit card fraud, targets are easily available because if your credit card gets stolen, you will report to the credit card company pretty soon that it has been stolen and so the credit card company can tag your account as being in a fraudulent state whereas in other industries where we are trying to use predictive analytics now say for example tax under filing, the tax under filers are not going to report that, “Hey, I am under filing my taxes this year.” So, you do not have that target information. Also, to take things further if you are… I have also worked in the area of network intrusion detection where you are monitoring a computer network, you don’t know which particular activity is malicious and you know which one is normal. So, again, you do not know the targets. After the fact, using forensics, you can identify some of the activities as being malicious or not, but you still are not sure you know whether you caught all the malicious activity or not and then going even one step further for bio surveillance; fortunately, we do not have any examples of bioterrorist attacks and so when you are building models, when you are trying to use predictive analytics for something like bioterrorism or bio surveillance, it is a much more of a challenging problem.

Bill Cullifer: Thank you. I certainly appreciate your perspective and for your time today.

Vijay Desai: Most welcome.

{ 0 comments }

 
icon for podpress  Maximizing Marketing Impact with Net Lift Models-Interview with Kim Larsen, Director of Advanced Analytics at Charles Schwab & Co (3 of 3): Play Now | Play in Popup | Download

In this third and final interview of Kim Larsen, Director of Advanced Analytics at Charles Schwab & Co. Kim shares his thoughts regarding strategies tp Maximize Marketing Impact with Net Lift Models.

Net Lift Models are designed to maximize the incremental impact by targeting the undecided clients that can be motivated by marketing.

Check out the three minute interview on the Predictive Analyticswebsite.

Transcript:

Kim Larson: That approach is my own sort of favorite because it’s very easy to do and it’s very intuitive. It’s using a k-NN, it’s like a nearest neighborhood classifier. So, it’s very simple essentially. First you identify say 10 or 20 variables that you think are great net lift variables meaning that they can differentiate conversion rates between test and control group and you can think of many ways to do that, but that’s actually quite easy for a single variable to see is this variable really a variable that can differentiate the self selectors from the… just the three other converters. Then, once you have… let’s say you got about 20 variables or so; based on those 20 variables, you find… you would go to a client, you will find say the 100 clients that look most like that client. So, you will basically find the 100 closest neighbors to that client like the “friends” of that client and then we are in that neighborhood of a 100 people, you basically just do a voting so to speak. You calculate the net lift weight for those 100 people and that net lift weight is the estimated net conversion rate for that client and then so forth. You can use SAS to do this or any other software that provides a k-NN classifier. In SAS, you can use [Inaudible] and I am sure about [Inaudible] and all those other packages have similar… [Voice Cross Over] it’s very easy to do, that’s the last one.

Bill Cullifer: Great.

Kim Larsen: Tends to give good results as well.

Bill Cullifer: Question for you. Is this a function of a business unit or information technology or of combination of both?

Kim Larsen: The LIFT model?

Bill Cullifer: Yeah, I mean, LIFT Model.

Kim Larsen: Yeah. I think typically you would… these models would be undertaken by some analytical team within your company like the team I worked in. I think typically most companies will have a modeling team and they would be the ones doing that. They would need to work with the marketing people to implement this into a campaign and the biggest challenge there is when you have got… when you have got a net model and you are telling people, “Well, these are the clients that I intend to target.” People will notice that these are not the typical people that you target. What I always tell people is that if you forget about all the fancy modeling and all the fancy stuff, there is a general rule of thumb that I have found. Your most engaged, wealthy, happy clients that love you the most, they are not the ones you need to hit with direct marketing. They are going to buy your stuff regardless, because they already like you, they already think you provide great service. Where you need to put your money is on the fence sitters, the people that are right below your best clients. Those are the people that are sleeping a little bit. They have all the propensity and the wealth and all that’s needed to really… to drive your bottom line, but they are not doing it fully because something… maybe they are not engaged with you. They don’t know what you offer, etc. So, that’s a general rule of thumb. Now, you as the analytical manager or the analytical analyst, you need to convince your marketing colleagues that that’s the way to go and that sometimes… I mean at Schwab it’s never really been an issue for me, but I know some other places that that can be an issue.

Bill Cullifer: Yeah, fair enough, I bet it can be.

Kim Larsen: Because you think…

Bill Cullifer: Then it’s an huge issue, right?

Kim Larsen: Yeah.

Bill Cullifer: And part of it is an educational process.

Kim Larsen: Exactly.

Bill Cullifer: Yeah, fair enough. Well, that’s an excellent perspective and I appreciate you sharing that as well and for your time today.

Kim Larsen: You are welcome.

{ 0 comments }

 
icon for podpress  Flash Video: Play Now | Play in Popup | Download

In this second in a series of interviews (2-3) with Kim Larsen, Director of Advanced Analytics at Charles Schwab & Co Kim shares his thoughts regarding strategies tp Maximize Marketing Impact with Net Lift Models. Net Lift Models are designed to maximize the incremental impact by targeting the undecided clients that can be motivated by marketing.

Check out the three minute interview on the Predictive Analyticswebsite.

A full transcript to follow in forty-eight hours.

{ Comments on this entry are closed }