Study Archives - Business Intelligence Info

Tag Archives: Study

University of Michigan study advocates ban of facial recognition in schools

August 11, 2020 Big Data

A newly published study by University of Michigan researchers shows facial recognition technology in schools presents multiple problems and has limited efficacy. Led by Shobita Parthasarathy, director of the university’s Science, Technology, and Public Policy (STPP) program, the research says the technology isn’t suited for security purposes and can actively promote racial discrimination, normalize surveillance, and erode privacy while institutionalizing inaccuracy and marginalizing non-conforming students.

The study follows the New York legislature’s passage of a moratorium on the use of facial recognition and other forms of biometric identification in schools until 2022. The bill, which came in response to the launch of facial recognition by the Lockport City School District, was among the first in the nation to explicitly regulate or ban use of the technology in schools. That development came after companies including Amazon, IBM, and Microsoft halted or ended the sale of facial recognition products in response to the first wave of Black Lives Matter protests in the U.S.

The University of Michigan study — a part of STPP’s Technology Assessment Project — employs an analogical case comparison method to look at previous uses of security technology (CCTV cameras, metal detectors, and biometric technologies) and anticipate the implications of facial recognition. While its conclusions aren’t novel, it takes a strong stance against commercial products it asserts could harm students and educators far more than it helps them.

For instance, the coauthors claim that facial recognition would disproportionately target and discriminate against people of color, particularly Black and Latinx communities. At the same time, they say that facial recognition would create new rules for dress and appearance and punish students who don’t fit into narrow standards of acceptability, causing problems whenever a school relies on it to automate activities like taking attendance or purchasing lunch.

Indeed, countless studies have shown facial recognition to be susceptible to bias. A paper last fall by University of Colorado, Boulder researchers showed that AI from Amazon, Clarifai, Microsoft, and others maintained accuracy rates above 95% for cisgender men and women but misidentified trans men as women 38% of the time. Separate benchmarks of major vendors’ systems by the Gender Shades project and the National Institute of Standards and Technology (NIST) suggest that facial recognition technology exhibits racial and gender bias and facial recognition programs can be wildly inaccurate, misclassifying people upwards of 96% of the time.

Facial recognition will take existing racial biases and make them worse, causing more surveillance and humiliation of Black and brown students, the University of Michigan researchers argue. And it will make surveillance a part of everyday life, laying the groundwork for expansion to other uses. Lockport portends this — while the schools’ privacy policy stated that its facial recognition watchlist wouldn’t include students and would only cover non-students deemed a threat, the district superintendent ultimately oversaw which individuals were added to the system and the school board president couldn’t guarantee student photos would never be included for disciplinary reasons.

The University of Michigan study’s coauthors also maintain that facial recognition in schools will create new kinds of student data that will be sold and bought by private corporations. Data collected for one purpose will be used in other ways, such that it will become impossible for students to provide full and informed consent to data collection or control. A legal remedy to this was proposed last week by Sen. Jeff Merkley (D-OR) and Sen. Bernie Sanders (I-VT) in the National Biometric Information Privacy Act, which would make it illegal for businesses to collect, purchase, or trade biometric information obtained from customers without permission. But few protections exist in most U.S. states as of now.

For these reasons, the researchers recommend a nationwide ban on facial recognition in schools. However, they provide policy recommendations for schools that deem the technology “absolutely necessary.” Among other steps, they propose a five-year moratorium on the use of facial recognition technology in schools; convening a national advisory committee to investigate facial recognition and its implications; establishing technology offices to help schools navigate the technical, social, ethical, and racial challenges of facial recognition; and deleting facial recognition data at the end of each academic year or when students graduate or leave the district.

A number of efforts to use facial recognition systems within schools have been met with resistance from parents, students, alumni, community members, and lawmakers alike. At the college level, a media firestorm erupted after a University of Colorado professor was revealed to have secretly photographed thousands of students, employees, and visitors on public sidewalks for a military anti-terrorism project. University of California San Diego researchers admitted to studying footage of students’ facial expressions to predict engagement levels. And last year, the University of California Los Angeles proposed using facial recognition software for security surveillance as part of a larger campus security policy.

Let’s block ads! (Why?)

Big Data – VentureBeat

NIST study finds that masks defeat most facial recognition algorithms

July 28, 2020 Big Data

In a report published today by the National Institutes of Science and Technology (NIST), a physical sciences laboratory and non-regulatory agency of the U.S. Department of Commerce, researchers attempted to evaluate the performance of facial recognition algorithms on faces partially covered by protective masks. They report that the 89 commercial facial recognition algorithms from Panasonic, Canon, Tencent, and others they tested had error rates between 5% and 50% in matching digitally applied masks with photos of the same person without a mask.

“With the arrival of the pandemic, we need to understand how face recognition technology deals with masked faces,” Mei Ngan, a NIST computer scientist and a coauthor of the report, said in a statement. “We have begun by focusing on how an algorithm developed before the pandemic might be affected by subjects wearing face masks. Later this summer, we plan to test the accuracy of algorithms that were intentionally developed with masked faces in mind.”

The study — part of a series from NIST’s Face Recognition Vendor Test (FRVT) program conducted in collaboration with the Department of Homeland Security’s Science and Technology Directorate, the Office of Biometric Identity Management, and Customs and Border Protection — explored how well each of the algorithms was able to perform “one-to-one” matching, where a photo is compared with a different photo of the same person. (NIST notes this sort of technique is often used in smartphone unlocking and passport identity verification systems.) The team applied the algorithms to a set of about 6 million photos used in previous FRVT studies, but they didn’t test “one-to-many” matching, which is used to determine whether a person in a photo matches any in a database of known images.

Because real-world masks differ, the researchers came up with nine mask variants to test, which included differences in shape, color, and nose coverage. The digital masks were black or a light blue approximately the same color as a blue surgical mask, while the shapes ranged from round masks covering the nose and mouth to a type as wide as the wearer’s face. The wider masks had high, medium, and low variants that covered the nose to varying degrees.

According to the researchers, algorithm accuracy with masked faces declined “substantially” across the board. Using unmasked images, the most accurate algorithms failed to authenticate a person about 0.3% of the time, and masked images raised even these top algorithms’ failure rate to about 5%, while many “otherwise competent” algorithms failed between 20% and 50% of the time.

In addition, masked images more frequently caused algorithms to be unable to process a face, meaning they couldn’t extract features well enough to make an effective comparison. The more of the nose a mask covered, the lower the algorithm’s accuracy; accuracy degraded with greater nose coverage. Error rates were generally lower with round masks and black masks as opposed to surgical blue ones. And while false negatives increased, false positives remained stable or modestly declined. (A false negative indicates an algorithm failed to match two photos of the same person, while a false positive indicates it incorrectly identified a match between photos of two different people.)

“With respect to accuracy with face masks, we expect the technology to continue to improve,” continued Ngan. “But the data we’ve taken so far underscores one of the ideas common to previous FRVT tests: Individual algorithms perform differently. Users should get to know the algorithm they are using thoroughly and test its performance in their own work environment.”

The results of the study align with a VentureBeat article earlier this year that found that facial recognition algorithms used by Google and Apple struggled to recognize mask-wearing users. But crucially, NIST didn’t take into account systems designed specifically to identify mask wearers, like those from Chinese company Hanwang and researchers affiliated with Wuhan University. In an op-ed in April, Northeastern University professor Woodrow Hartzog characterized masks as a temporary technological speed bump that won’t stand in the way of increased facial recognition use in the age of COVID-19. Already, companies like Clearview AI are attempting to sell facial recognition to state agencies for the purpose of tracking people infected with COVID-19.

Perhaps in recognition of this, this summer, NIST plans to test algorithms created with face masks in mind and conduct tests with one-to-many searches and other variations.

Let’s block ads! (Why?)

Big Data – VentureBeat

‘Fundamentally flawed’ study describes facial recognition system designed to identify non-binary people

July 15, 2020 Big Data

Last Chance: Register for Transform, VB’s AI event of the year, hosted online July 15-17.

In a paper published on the preprint server Arxiv.org, coauthors affiliated with Harvard and Autodesk propose extending current facial recognition systems’ capabilities to identify “gender minority subgroups” such as the LGBTQ and non-binary communities. They claim the corpora they created — a “racially balanced” database capturing a subset of LGBTQ people and an “inclusive-gender” database — can mitigate bias in gender classification algorithms. But according to the University of Washington AI researcher Os Keyes, who wasn’t involved with the research, the paper appears to conceive of gender in a way that’s not only contradictory, but dangerous.

“The researchers go back and forth between treating gender as physiologically and visually modeled in a fixed way, and being more flexible and contextual,” Keyes said. “I don’t know the researchers’ backgrounds, but I’m at best skeptical that they ever spoke to trans people about this project.”

Facial recognition is problematic on its face — so much so that the Association for Computing Machinery (ACM) and American Civil Liberties Union (ACLU) continue to call for moratoriums on all forms of it. (San Francisco, Oakland, Boston, and five other Massachusetts communities have banned the use of facial recognition by local departments, and after the height of the recent Black Lives Matter protests in the U.S., companies including Amazon, IBM, and Microsoft halted or ended the sale of facial recognition products.) Benchmarks of major vendors’ systems by the Gender Shades project and the National Institute of Standards and Technology (NIST) have found that facial recognition exhibits race, gender bias, and poor performance on people who don’t conform to a single gender identity. And facial recognition can be wildly inaccurate, misclassifying people upwards of 96% of the time.

In spite of this, the paper’s coauthors — perhaps with the best of intentions — sought to improve the performance of facial recognition systems when they’re applied to transgender and non-binary people. They posit that current facial recognition algorithms are likely to amplify societal gender bias and that the lack of LGBTQ representation in popular benchmark databases leads to a “false sense of progress” on gender classification tasks in machine learning, potentially harming the self-confidence and psychology of those misgendered by the algorithms.

That’s reasonable, according to Keyes, but the researchers’ assumptions about gender are not.

“They settle on treating gender as fixed, and modeling non-binary people as a ‘third gender’ category in between men and women, which isn’t what non-binary means at all,” Keyes said. “People can be non-binary and present in very different ways, identify in very different ways, [and] have many different life histories and trajectories and desired forms of treatment.”

Equally problematic is that the researchers cite and draw support from a controversial study implying all gender transformation procedures, including hormone replacement therapy (HRT), cause “significant” face variations over time, both in shape and texture. Advocacy groups like GLAAD and the Human Rights Campaign have denounced the study as “junk science” that “threatens the safety and privacy of LGBTQ and non-LGBTQ people alike.”

“This junk science … draws on a lot of (frankly, creepy) evolutionary biology and sexology studies that treat queerness as originating in ‘too much’ or ‘not enough’ testosterone in the womb,” Keyes said. “Again, those studies haven’t been validated — they’re attractive because they imply that gay people are too feminine, or lesbians too masculine, and reinforce social stereotypes. Depending on them and endorsing them in a study the authors claim is for mitigating discrimination is absolutely bewildering.”

The first of the researchers’ databases — the “inclusive database” — contains 12,000 images of 168 unique identities, including 29 white males, 25 white females, 23 Asian males, 23 Asian females, 33 African males, and 35 African females from different geographic regions, 21 of whom (9% of the database) identify as LGBTQ. The second — the non-binary gender benchmark database — comprises 2,000 headshots of 67 public figures labeled as “non-binary” on Wikipedia.

Keyes takes issue with the second data set, which they argue is non-representative because it’s self-selecting and because of the way appearance tends to be policed in celebrity culture. “People of color, disabled people, poor people need not apply — certainly not as frequently,” Keyes said. “It’s sort of akin to fixing bias against women by adding a data set exclusively of women with pigtails; even if it ‘works,’ it’s probably of little use to anyone who doesn’t fit a very narrow range of appearances.”

The researchers trained several image classification algorithms on a “racially-imbalanced” but popular facial image database — the Open University of Israel’s Adience — augmented with images from their own data sets (1,500 images from the inclusive database and 1,019 images from the non-binary database). They then applied various machine learning techniques to mitigate algorithmic bias and boost the models’ accuracy, which they claim enabled the best-performing model to predict non-binary people with 91.97% accuracy.

The results in any case ignore the fact that “trans-inclusive” systems for non-consensually defining someone’s gender are a contradiction in terms, according to Keyes. “When you have a technology that is built on the idea that how people look determines, rigidly, how you should classify and treat them, there’s absolutely no space for queerness,” they said. “Rather than making gender recognition systems just, or fair, what projects like this really do is provide a veneer of inclusion that serves mostly to legitimize the surveillance systems being built — indeed, it’s of no surprise to me that the authors end by suggesting that, if there are problems with their models, they can be fixed by gathering more data; by surveilling more non-binary people.”

Let’s block ads! (Why?)

Big Data – VentureBeat

The Columbia COVID Study

May 21, 2020 Humor

Really interesting study about COVID-19 out of Columbia University. This study based their model on actual data at the county level (rather than state or even country), which is important because virus transmission is always local, and used 7-day averages, which is important to avoid data noise.

The result that is getting all the headlines is that if the US had started distancing measures just 1 week earlier than we did, then the death toll would have been reduced 55% (less than half the people would have died). And if the US had started 2 weeks earlier, then they would have been reduced by 83%.

This is all about exponential growth. Small changes early on lead to huge changes in results later.

A study result that is even more interesting to me is that they actually showed that when distancing measures are relaxed, there is a significant delay in response. Here’s the important quote from the study:

… a decline of daily confirmed cases continues for almost two weeks after easing of control measures. … This decreasing trend, caused by the NPIs [non-pharmaceutical interventions] in place prior to May 4, 2020 coupled with the lag between infection acquisition and case confirmation, conveys a false signal that the pandemic is well under control. Unfortunately, due to high remaining population susceptibility, a large resurgence of both cases and deaths follows.

The original study itself is a bit of a slog. The NY Times has a easier-to-read summary. If you don’t have access to that, I’m sure there will be more articles about this soon (although the further away you get from the original study the more it gets muddied by the media).

Bottom line:

Timing is everything in the spread of a virus. If you wait until things are bad, it is way too late. Just a one-week delay killed 36,000 people by May 3 (more than ten times the death toll from 9/11). It will continue to kill more in the future if we don’t get this under control.
Opening things up and then waiting two weeks to see what happens is a trap. We are already falling for that trap. If we fall for it a second time, then shame on us.

Let’s block ads! (Why?)

Political Irony

Google’s ML-fairness-gym lets researchers study the long-term effects of AI’s decisions

February 6, 2020 Big Data

Determining whether an AI system is maintaining fairness in its predictions requires an understanding of models’ short- and long-term effects, which might be informed by disparities in error metrics on a number of static data sets. In some cases, it’s necessary to consider the context in which the AI system operates in addition to the aforementioned error metrics, which is why Google researchers developed ML-fairness-gym, a set of components for evaluating algorithmic fairness in simulated social environments.

ML-fairness-gym — which was published in open source on Github this week –is designed to be used to research the long-term effects of automated systems by simulating decision-making using OpenAI’s Gym framework. AI-controlled agents interact with digital environments in a loop, and at each step, an agent chooses an action that affects the environment’s state. The environment then reveals an observation that the agent uses to inform its next actions, so that the environment models the system and dynamics of a problem and the observations serve as data.

For instance, given the classic lending problem, where the probability that groups of applicants pay back a bank loan is a function of their credit score, the bank acts as the agent and receives applicants, their scores, and their membership in the form of environmental observations. It makes a decision — accepting or rejecting a loan — and the environment models whether the applicant successfully repays or defaults and adjusts their credit score accordingly. Throughout, ML-fairness-gym simulates the outcomes so that the effects of the bank’s policies on fairness to the applicants can be assessed.

ML-fairness-gym in this way cleverly avoids the pitfalls of static data set analysis. If the test sets (i.e., corpora used to evaluate model performance) in classical fairness evaluations are generated from existing systems, they may be incomplete or reflect the biases inherent to those systems. Furthermore, the actions informed by the output of AI systems can have effects that might influence their future input.

Google’s ML fairness gym lets researchers study the long term effects of AI’s decisions

Above: In the lending problem scenario, this graph illustrates changing credit score distributions for two groups over 100 steps of simulation.

Image Credit: Google

“We created the ML-fairness-gym framework to help ML practitioners bring simulation-based analysis to their ML systems, an approach that has proven effective in many fields for analyzing dynamic systems where closed form analysis is difficult,” wrote Google Research software engineer Hansa Srinivasan in a blog post.

Several environments that simulate the repercussions of different automated decisions are available, including one for college admissions, lending, attention allocation, and infectious disease. (The ML-fairness-gym team cautions that the environments aren’t meant to be hyper-realistic and that best-performing policies won’t necessarily translate to the real world.) Each have a set of experiments corresponding to published papers, which are meant to provide examples of ways ML-fairness-gym can be used to investigate outcomes.

The researchers recommend using ML-fairness-gym to explore phenomena like censoring in the observation mechanism, errors from the learning algorithm, and interactions between the decision policy and the environment. The simulations allow for the auditing of agents to assess the fairness of decision policies based on observed data, which can motivate data collection policies. And they can be used in concert with reinforcement learning algorithms — algorithms that spur on agents with rewards — to derive new policies with potentially novel fairness properties.

In recent months, a number of corporations, government agencies, and independent researchers have made attempts at tackling the so-called “black box” problem in AI — the opaqueness of some AI systems — with varying degrees of success.

“Machine learning systems have been increasingly deployed to aid in high-impact decision-making, such as determining criminal sentencing, child welfare assessments, who receives medical attention and many other settings,” continued Srinivasan. “We’re excited about the potential of the ML-fairness-gym to help other researchers and machine learning developers better understand the effects that machine learning algorithms have on our society, and to inform the development of more responsible and fair machine learning systems.”

In 2017, the U.S. Defense Advanced Research Projects Agency launched DARPA XAI, a program that aims to produce “glass box” models that can be easily understood without sacrificing performance. In August, scientists from IBM proposed a “factsheet” for AI that would provide information about a model’s vulnerabilities, bias, susceptibility to adversarial attacks, and other characteristics. A recent Boston University study proposed a framework to improve AI fairness. And Microsoft, IBM, Accenture, and Facebook have developed automated tools to detect and mitigate bias in AI algorithms.

Let’s block ads! (Why?)

Big Data – VentureBeat

Generational In-store Retail Preferences not What You Might Think, Study Finds

June 30, 2019 NetSuite

Posted by Barney Beal, Content Director

Today’s retailers, already contending with evolving customer preferences, aging IT infrastructure poorly suited to adapt to modern demands, and competition not only from Amazon but also from manufacturers and distributors now face a new challenge—meeting customer expectations across generations.

A recent study conducted by Oracle NetSuite, Wakefield Research and The Retail Doctor, found some significant differences in expectations, and also challenged a few stereotypes across four generations: baby boomers, Gen X, millennials and Gen Z.

The study surveyed 1,200 consumers and 400 retail executives across the U.S., U.K. and Australia.

Generations Have Different Expectations for What’s In-Store

Some of the most profound, and most surprising, differences appear in the in-store shopping experience. Know who’s more likely to do more in-store shopping this year? No, not the baby boomers (13%) and Gen X (29%). It’s the “digital native” Gen Z and millennials (43%), according to the study. More of the younger generations view the retail experience positively (57%) than Gen X (40%) and baby boomers (13%).

The results do not suggest, however, that retailers can sit back and revel in the fact that they have appealed to the coveted younger demographic with the in-store experience.

First, there’s a case to be made that young people still like going out shopping with their friends, whether they’re digitally literate, smartphone users or not. Moreover, when they do get in the store, generations view their interactions with staff very differently. Among Gen Z, 42% are more annoyed by increased interaction by retail associates. Millennials (56%), Gen X (44%) and baby boomers (43%) all noted they would feel more welcomed by more in-store interactions.

What’s the takeaway for retailers? Clearly, Gen Z is the first truly “digital native” generation. They know what they’re looking for, they’ve often researched a purchase before they enter the store and, when they do need help, they expect the store associate to know more than they do. They also want an associate that isn’t pushing products that aren’t relevant. That means store associates need to be sufficiently trained and equipped to handle any situation that arises, whether that’s recommending complementary products or, if something is out of stock, shipping it to the customer’s home from a warehouse or even a nearby store. Provide a poor experience and the customer is likely to take their complaints to social media—something true of all generations.

Invest in Social Media, with Tempered Expectations

However, social media is not the panacea many retailers believe it to be, according to the study. While almost all retail executives (98%) think that engaging customers on social media is important to building stronger relationships, overall only 12% of consumers think social media has a significant impact on the way they think or feel about a brand.

For those that do engage with brands over social media, the results more closely follow expectations with Gen Z consumers (38%) much more likely than other generations to engage with retailers on social to get to know the brand compared to millennials (25%), Gen X (27 %) and baby boomers (21%). Similarly, Gen Z (65%) consumers and millennials (62%) believe social media platforms have an impact on their relationship with brands, while more than half of baby boomers (53%) and 29% of Gen X consumers do not engage with brands on social media.

The takeaway here for retailers? Those that want to reach millennials and Gen Z had better make a commitment to social. Social media is a form of trust; those that are not active on social channels are viewed as less trustworthy. For example, Gen Z expects brands to showcase their personality online.

If a retailer wants to connect and influence purchases, they need to be active and responsive on social. If a retailer isn’t, their younger customers are. Many Gen Z apparel customers post pictures of items they are considering and then 46% make a purchase based on the feedback (social proof) to that post.

Access the full report for more details on generational differences in retail expectations and what retail executives are thinking.

Posted on Tue, June 25, 2019
by NetSuite filed under

Let’s block ads! (Why?)

The NetSuite Blog

New study: “Digital natives” value brick and mortar stores more than their parents or grandparents

June 26, 2019 NetSuite

Global Study Highlights the Varying Shopping Expectations of Different Generations and the Role of Technology in Personalizing Retail

REDWOOD CITY, CA.—June 25, 2019—Despite clear differences in expectations among shoppers of different generations, almost half of retailers (44 percent) have made no progress in tailoring the in-store shopping experience according to a recent study conducted by Oracle NetSuite, Wakefield Research and The Retail Doctor. The global study of 1,200 consumers and 400 retail executives across the U.S., U.K. and Australia dispelled stereotypes around generations and found big differences in generational expectations across baby boomers, Gen X, millennials and Gen Z.

“We have seen decades of diminishing experiences in brick and mortar stores, and the differences identified in these results point to its impact on consumers over the years,” said Bob Phibbs, CEO, The Retail Doctor. “Retailers have fallen behind in offering in-store experiences that balance personalization and customer service but there’s an opportunity to take the reins back. The expectation from consumers is clear and it’s up to retailers to offer engaging and custom experiences that will cater to shoppers across a diverse group of generations.”

Beauty is in the eye of the beholder: Retailers struggle to keep stride with generational shoppers

The in-store shopping experience remains an important part of the retail environment for all generations, but the progress retailers are making to improve the in-store experience is being viewed differently by different generations.

Despite the stereotypes of “digital natives”, Gen Z and millennials (43 percent) are most likely to do more in-store shopping this year followed by Gen X (29 percent) and baby boomers (13 percent).
Gen Z and millennials (57 percent) had the most positive view of the current retail environment feeling it was more inviting, followed by Gen X (40 percent). Baby boomers (27 percent) were more likely to find the current retail environment less inviting than consumers overall.
Gen Z valued in-store interaction the least with 42 percent feeling more annoyed from increased interaction with retail associates. In contrast, millennials (56 percent), Gen X (44 percent) and baby boomer (43 percent) generations all noted they would feel more welcomed by more in-store interactions.

Retailers view emerging technologies through rose-colored glasses

While more than three quarters of retail executives (79 percent) believe having AI and VR in stores will increase sales, the study found that these technologies are not yet widely accepted by any generation.

Overall, only 14 percent of consumers believe that emerging technologies like AI and VR will have a significant impact on their purchase decisions.
Emerging tech in retail stores is most attractive to millennials (50 percent) followed by Gen Z (38 percent), Gen X (35 percent) and baby boomers (20 percent).
Perceptions of VR varied widely across different generations. Fifty-eight percent of Gen Z said VR would have some influence on their purchase decisions, while 59 percent of baby boomers said VR would have no influence on their purchase decision.

Insta-famous brands reach Gen Z and millennial consumers, but not as much as retailers think

While almost all retail executives (98 percent) think that engaging customers on social media is important to building stronger relationships with them, the study found a big disconnect with consumers across all generations.

Overall, only 12 percent of consumers think their engagement with brands on social media has a significant impact on the way they think or feel about a brand.
Among those who engage with brands on social media, Gen Z (38 percent) consumers are much more likely than other generations to engage with retailers on social to get to know the brand compared to millennials (25 percent) and baby boomers (21 percent).
Gen Z (65 percent) consumers and millennials (63 percent) believe their engagement with brands on social media platforms have an impact on their relationship with brands.
More than half of baby boomers (53 percent) and 29 percent of Gen X consumers do not engage with brands on social media.

“After all the talk about brick and mortar stores being dead, it’s interesting to see that ‘digital natives’ are more likely to increase their shopping in physical stores this year than any other generation,” said Greg Zakowicz, senior commerce marketing analyst, Oracle NetSuite. “Stepping back, these findings fit with broader trends we have been seeing around the importance of immediacy and underlines why retailers cannot afford to make assumptions about the needs and expectations of different generations. It really is a complex puzzle and as this study clearly shows, retailers need to think carefully about how they meet the needs of different generations.”

To read more about NetSuite’s insights into the report’s finding visit NetSuite’s cloud blog.

Methodology
For this survey, 1,200 consumers and 400 retail executives were surveyed around the overall retail environment, in-store and online shopping experiences and advanced technologies. Both retailers and consumers were surveyed from three global markets including the U.S., U.K. and Australia with retail executives representing organizations between $ 10-100 million in annual sales.

About Wakefield Research
Wakefield is a full-service market research firm that uncovers insights for brands to help them solve problems and grow their business. Wakefield Research is a partner to the world’s leading consumer and B2B brands, including 50 of the Fortune 100. Wakefield Research conducts qualitative and quantitative research in 70 countries. For more information, please visit https://www.wakefieldresearch.com

About The Retail Doctor
The Retail Doctor is a New York-based retail consulting firm created by expert retail consultant and leading business mentor Bob Phibbs. With over 30 years of experience in retail, Bob has worked as a consultant, speaker, and entrepreneur, helping businesses revolutionize their brand and grow their success. Bob is also the author of three highly-praised books, including The Retail Doctor’s Guide to Growing Your Business (WILEY). His clients include some of the largest retail brands in the world including Bernina, Brother, Caesars Palace, Hunter Douglas, Lego, Omega and Yamaha. For more information, please visit www.retaildoc.com

About Oracle NetSuite
For more than 20 years, Oracle NetSuite has helped organizations grow, scale and adapt to change. NetSuite provides a suite of cloud-based applications, which includes financials / Enterprise Resource Planning (ERP), HR, professional services automation and omnichannel commerce, used by more than 18,000 customers in 203 countries and dependent territories.

For more information, please visit http://www.netsuite.com.

Follow NetSuite’s Cloud blog, Facebook page and @NetSuite Twitter handle for real-time updates.

About Oracle
The Oracle Cloud offers complete SaaS application suites for ERP, HCM and CX, plus best-in-class database Platform as a Service (PaaS) and Infrastructure as a Service (IaaS) from data centers throughout the Americas, Europe and Asia. For more information about Oracle (NYSE:ORCL), please visit us at oracle.com.

Trademarks
Oracle and Java are registered trademarks of Oracle and/or its affiliates.

Safe Harbor
The preceding is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release and timing of any features or functionality described for Oracle’s products remains at the sole discretion of Oracle.

Let’s block ads! (Why?)

NetSuite's Latest Press Coverage

Facebook AI study: Major object recognition systems favor people with more money

June 9, 2019 Big Data

Computer vision for recognizing household objects works better for people in high-income households, according to analysis of 6 major object detection systems shared today by Facebook AI researchers. The study examined object classification systems made by Facebook, Google Cloud, Microsoft Azure, AWS, IBM Watson, and Clarifai.

Results show the 6 systems work between 10-20% better for the wealthiest households than they do for the poorest households.

A company spokesperson declined to share specific figures on the performance of other individual companies, but Facebook’s system had an accuracy gap as high as 20% between a home making $ 3,500 a month or more in the United States and a household with an income of $ 50 a month or less in countries like Somalia and Burkina Faso.

Facebook AI study: Major object recognition systems favor people with more money

Above: Aggregate object recognition system performance according to Facebook

The systems were generally more likely to identify items in homes in North America and Europe than in Asia and Africa. Results are detailed in a paper published by researchers on arXiv Thursday titled “Does Object Recognition Work for Everyone?”

Object recognition uses computer vision to discern the difference between things like a chair or toothpaste or a dress. Object recognition is harnessed by a number of cloud service companies as well as for consumer-facing services like Google Assistant’s computer vision service Lens and Amazon’s StyleSense. Facebook uses object detection for things like content moderation and recognizing items on the screen for people with visual impairments.

Above: Combined object recognition performance results

“Our analysis showed that this issue is not specific to one particular object-recognition system, but rather broadly affects tools from a wide range of companies, including ours,” reads a blog post announcing the news today. “These results clearly show that we must do better both across the industry and here at Facebook.”

A map of Facebook’s object detection performance finds it performs worst in the southern hemisphere.

Above: Facebook object detection accuracy map

“By publishing these results and describing our methodology, AI researchers and engineers across the community can use this work to test and compare the performance of their own object-recognition systems, and then make them more capable of serving everyone effectively,” the post read.

To test the accuracy of object recognition systems, researchers used Dollar Street, an open source data set compiled by the Gapminder Foundation of images taken from 264 homes in 50 countries.

The discrepancy in the major systems may be the result of the fact that data sets like ImageNet used to train many object detection systems are compiled almost entirely from photos in Europe and North America, researchers said.

Photos obtained through English-language searches on public photo websites could also be a source of the disparity in the overrepresentation of higher income environments in the U.S. and Europe.

Facebook said it plans to address the shortcoming by training its convolutional networks with images that contain hashtags in languages other than English.

The work will be shared at the Computer Vision for Global Challenges workshop being held at the Computer Vision and Pattern Recognition (CVPR) conference, which takes place June 16-20 in Long Beach, California.

Today’s announcement follows Facebook’s presentation last month detailing how Facebook tests its computer vision and AR systems to ensure they work on people with different skin tones.

In related news, Facebook computer vision researchers had to redefine the kinds of buildings that should be labeled a home when making a population density map of Africa earlier this year.

Let’s block ads! (Why?)

Big Data – VentureBeat

AI Weekly: AI is changing the way we study the stars, grow food, and create art

April 13, 2019 Big Data

Too often, technologists become wrapped up in doom-and-gloom predictions about job-stealing, prejudicial, and potentially murderous AI. Fear sells, the saying goes, and that seems doubly true when it comes to emerging tech.

But focusing on AI’s negatives blinds us to its positives. As my colleague Khari Johnson and I have written countless times, artificial intelligence promises to transform entire verticals for the better, from health care and education to business intelligence and cybersecurity. More excitingly, it’s laying the groundwork for new industries and pursuits of which we haven’t yet conceived.

This week, MIT graduate student and postdoctoral fellow with Event Horizon Telescope Katie Bouman created an algorithm — Continuous High-resolution Image Reconstruction using Patch priors, or CHIRP for short — that combined data from eight radio telescopes from around the globe to generate the first image ever of a black hole. CHIRP — a three-year collaborative effort among MIT’s Computer Science and Artificial Intelligence Laboratory, the Harvard-Smithsonian Center for Astrophysics, and the MIT Haystack Observatory — reconstructs images while accounting for variations in signal strength, such that delays caused by atmospheric noise cancel each other out.

In equally uplifting news, the University of California, Berkeley on Monday unveiled a humanoid robot with a depth-sensing camera and motorized arms, all of which can be controlled with virtual reality handsets and trained to manipulate objects using AI. Pieter Abbeel, a professor and director of the Robot Learning Lab at UC Berkeley and the roboticist leading the project, told The Verge that recent advances in machine learning made possible the new design, which has a bill of materials substantially lower than most comparable alternatives (around $ 5,000 versus tens of thousands of dollars).

“The fact that AI is becoming more capable gave us an opportunity to rethink how to [build] a robot,” he said, adding that he expects robots of the future will be “reactive and dynamic.”

“There’s still a lot of challenges ahead, and it’s not like we think this specific robot is going in a home. [But] this is a design paradigm that takes us in a new direction,” Abbeel said.

AI isn’t just setting the stage for new robots and astronomical visualizations — it’s enabling scientists to grow more flavorful herbs, too. In a study published in the journal PLOS One earlier this month, scientists describe basil raised in climatic conditions that were optimized to increase the concentration of compounds responsible for its taste. A machine learning algorithm trained on data collected via mass spectrometry and gas chromatography yielded surprising insights, like that plants lit 24 hours a day taste superior to those which experience extended darkness.

Algorithms are transforming food in other ways. In February, IBM announced that it was teaming up with McCormick & Company to create new flavors and foods with machine learning. New York startup Analytical Flavor Systems’ platform taps sensory data and AI to suss out products’ flavor profiles and identify ways they might be improved. Meanwhile, companies like Los Angeles-based Halla are using AI to generate Netflix-like recommendations for grocery, restaurant, and food delivery apps and websites, in part by tapping databases of restaurant dish, recipe, ingredient, and grocery item taste and flavor attributes.

And in the art and science realms, AI is obviating mundane and repetitive tasks that take time away from the creative process. Disney Research recently detailed an algorithm that generates storyboards from scripts, letting creators visualize scenes before committing hours to animating them. Meanwhile, other researchers are experimenting with systems that can hallucinate foundational sketches of cats, fire trucks, mosquitos, and yoga poses, and companies like Promethean AI are employing machine learning to help human artists create art for video games. Yet another novel use case comes from academic publisher Springer Nature, which this week published a hefty research tome containing quotations, links, and references collated by AI.

That’s all to say that, for the many fears surrounding AI, there’s much good it can do — and already is doing. It’s important to keep that in mind in light of ongoing controversies.

For AI coverage, send news tips to Khari Johnson and Kyle Wiggers — and be sure to subscribe to the AI Weekly newsletter and bookmark our AI Channel.

Thanks for reading,

Kyle Wiggers

AI Staff Writer

Let’s block ads! (Why?)

Big Data – VentureBeat

New Study: Finance Is Set To Gain Digital Transformation Momentum

February 27, 2019 SAP

“Are we there yet?” That’s the question that is top of mind for architects of finance digital transformation. The answer is “not yet, but we’re getting much closer,” according to The Hackett Group’s key issues study, “2019 CFO Agenda: Building Next-Generation Capabilities.” Our findings offer a sobering yet optimistic reality check on just how far finance has progressed in its digital journey.

It’s tempting to assume that finance has advanced significantly, given all the hype and the rapid introduction of new technologies. However, our study found that the current impact of digital transformation on finance is limited. As the image below illustrates, only 31% of finance executives reported that digital transformation is currently having a high or very high impact on finance performance. Less than 30% have experienced a significant effect on their finance delivery model or ability to achieve enterprise objectives. Even the talent impact is muted.

This sounds worrisome. But if you take a look at the right side of the chart, you will see great cause for optimism. While the impact may be only partial today, it is expected to shoot up by an average of 2.6X in the next two to three years.

Feb 26 Figure 1 New Study: Finance Is Set To Gain Digital Transformation Momentum

This begs the question: How will finance realize greater digitalization benefits – that is, what’s going to change between today and two to three years from now? We found the answer by examining four critical enablers of digital change: the maturity and alignment of finance’s digital strategy, the function’s execution capabilities, and the level of support it gets from IT (see image below). While the data shows a low level of current enablement, it also portends a momentous rise across all four.

Feb 26 Figure 2 New Study: Finance Is Set To Gain Digital Transformation Momentum

Having a clear digital transformation roadmap is an essential success factor. The percentage of finance organizations with a mature strategy will double in the next two to three years. This time frame makes sense. We know from The Hackett’s Group Digital Transformation Performance Study (2018) that it takes digital leaders an average of 2.5 years to fully evolve their strategy. The same study revealed that 58% of finance organizations are currently in the process of developing their strategy. The speed of development will accelerate as early adopters share successful use cases.

It’s all in the execution

A great strategy is useless without having the right resources and competencies, including adequate IT support. The 2019 Key Issues Study shows that only 24% of finance organizations already have the execution capability and just 31% get the help they need from IT. As these gaps narrow dramatically, finance will rev up its digitalization engines. The digital transformation study found that:

Finance projects that the share of its IT budget dedicated to new initiatives will rise in the next 12 to 18 months.

Finance expects an increase in the percentage of project funding dedicated to digital efforts.

Finance anticipates that the share of FTEs dedicated to digital projects will rise, as well.

Of course, successful execution is not just about having more money and a bigger headcount. It’s equally about having people with the right skills. Our Digital Skills Poll (2018) revealed that typical finance organizations still suffer severe talent shortages in key areas. The most significant gaps between finance’s skills today and the skills it must master in a digital world are in data and technology savviness and analytical, innovative/creative, and design thinking. The good news is that the same poll also showed that the leaders in digital skills have already dramatically narrowed the gaps, at least by half, and in some cases even more than that. This means it is absolutely possible for finance to upskill, even as demand for talent continues to outstrip supply.

Conclusion

Finance organizations have been slower to show progress in digital transformation and realize its benefits compared to digital top performers across business services functions. That may sound bleak, but we have every indication that finance is gathering momentum. It’s poised to bolster its execution gap and more quickly develop a holistic strategy that is aligned with the enterprise’s overall digital transformation vision. Our research reveals substantial projected increases in the adoption rate of smart technologies like robotic process automation, cognitive computing, and next-generation enterprise resource planning solutions. Finance can catch up. And it must.

For more on the state of digital transformation in finance, see “Where Does Finance Stand Versus Digital Leaders?“

Let’s block ads! (Why?)

Digitalist Magazine

Tag Archives: Study

University of Michigan study advocates ban of facial recognition in schools

NIST study finds that masks defeat most facial recognition algorithms

‘Fundamentally flawed’ study describes facial recognition system designed to identify non-binary people

The Columbia COVID Study

Google’s ML-fairness-gym lets researchers study the long-term effects of AI’s decisions

Generational In-store Retail Preferences not What You Might Think, Study Finds

New study: “Digital natives” value brick and mortar stores more than their parents or grandparents

Facebook AI study: Major object recognition systems favor people with more money

AI Weekly: AI is changing the way we study the stars, grow food, and create art

New Study: Finance Is Set To Gain Digital Transformation Momentum

It’s all in the execution

Conclusion

Recent Posts

Categories

Archives