Your address will show here +12 34 56 78
Blog, eCommerce

Machine learning seems to be at the forefront of everyone’s mind. The buzz at NRF Retail’s BIG Show 2017 was no different. See what we learned about the intersection of retail and machine learning. Machine learning, combined with strong datasets and shoppers and trends, is already helping retailers create a customized, seamless, and predictive experience for their shoppers. As the predictive capabilities of computers improve, so will both online and […]

Machine learning seems to be at the forefront of everyone’s mind. The buzz at NRF Retail’s BIG Show 2017 was no different. See what we learned about the intersection of retail and machine learning.

Machine learning, combined with strong datasets and shoppers and trends, is already helping retailers create a customized, seamless, and predictive experience for their shoppers. As the predictive capabilities of computers improve, so will both online and offline shopping user experiences. With this advancement come opportunities for data scientists and more.

Data science: the key to personalization

One of the many machine learning technologies that we saw at NRF 2017 revolves around the idea of detailed personalization. As the amount of data on each individual shopper increases, so do the opportunities for data scientists to crunch the data. Algorithms that predict what users might like best, or what they might not yet know they like, are already in use and will continue to improve.

In another application, facial-recognition technology can be used on shoppers in brick-and-mortar stores to identify them, pull up the existing data on them, and guide them to the items that they are more likely to like and ultimately purchase. This kind of facial recognition technology can be used to reward loyal shoppers coming back for another visit. It is also bridging online and offline shopping by matching faces to online shopper personas and search or purchase history.


Testing out facial recognition technology that can tell gender, mood, and age range

New realities and interactive possibilities

Machine learning has interactive applications, too. Imagine being able to test-drive a new sofa, seeing exactly how it will look in your living room and debating whether it matches the color of your walls or rug. Retailers are eager to apply both virtual reality (VR) and augmented reality (AR). With VR experiences, virtual stores are places you can “visit” from the comfort of your home with the use of a headset. Imagine virtually browsing aisles of a grocery store without having to navigate through crowds with a shopping cart that inevitably has that one wobbly wheel.

On the other end of the [x]-reality spectrum is AR, which would help shoppers in the earlier sofa example. With AR, one can see both real “reality” and an augmented portion of it – in this case the potential sofa. This kind of technology would also allow shoppers to “try-on” clothes without working up a sweat shimmying into and out of too many pairs of jeans. At NRF we saw retailers using these new technologies to make shopping experiences more seamless and enjoyable for shoppers.

Virtual Reality
Using VR to design a new living room

On top of these, artificial intelligence can add yet another interactive layer. Chatbots – computer programs powered by artificial intelligence that are able to improvise and converse like humans – are being deployed online to chat with shoppers about what they’re looking for and help them find it. Though still in their infancy, and with examples to show that there is room for improvement, chatbots could be the future of everything from personalized shopping to customer service.

Taken together, machine learning’s many applications have a lot to offer to retail companies both online and offline.

But the implications of this growth go further still.

As dealers in data, we look at these trends and see a parallel rise in the need for massive data storage, processing, and analysis. Everyone from cloud technology architects to data scientists to data security experts should keep their eyes on the rapidly growing application of ML to retail and ecommerce.

The opportunities will be BIG!

0

Blog

Having more women in the room is powerful. Whether as engineers, board members, or VCs and more, increased gender diversity in technology is good ethics and good business. For example, increasing women’s participation on boards significantly improves financial performance. Women working as VCs increases start-up success and industry performance. Even small increases women’s workforce participation could boost the US’ economic output by $2.1 trillion, or by […]

Having more women in the room is powerful. Whether as engineers, board members, or VCs and more, increased gender diversity in technology is good ethics and good business.

For example, increasing women’s participation on boards significantly improves financial performance. Women working as VCs increases start-up success and industry performance. Even small increases women’s workforce participation could boost the US’ economic output by $2.1 trillion, or by over $12 trillion globally.

However, inequality in hiring practices and education continues.

Throughout my career, I’ve seen what this inequality looks like. I also know how powerful overcoming it can be.

While studying engineering, I was one of just 17 female engineers among almost 3000 male engineering students. At HP in its early days, I was the first female R&D engineer in my department, the first woman to go into the southern sales region, and the first female HP engineer in Germany. I was part of the team that started HP in India, in an era when people didn’t believe that IT could be done there.

Societal norms and family pressure kept suggesting to me – and my female colleagues – that “women can’t be engineers!” Together, we went on to defy these pressures and build new norms.

Along the way, I learned ways to encourage women in fields like engineering and technology. The best teachers I’ve had are the women at iMerit.

One of our delivery centers, located in a conservative Muslim neighborhood in Kolkata, India, is made up of over 90% women. Company-wide, women make up 60% of our employees.

These women work on advanced machine learning projects like building datasets for algorithms that power artificial intelligence. Here is some of what we have learned through them.

Train for the future

Our women show us that this inequality in access to technology education is not a factor of their interests or capabilities.

When access is provided, they engage with training opportunities: from computer literacy to understanding how machine learning works. Even those with little or no formal computer education excitedly seize training opportunities.

Addressing disparity in educational access is an important part of unlocking the potential of women. In the US, the percentage of women seeking computer science degrees has dropped over time. Even before college, girls pursue computer science at much lower rates than boys. In India, this pipeline problem is magnified by girls’ unequal access to any education.

By meeting these women where they are, and equipping them with future-proof skills, they can become the future shapers of technology.

Work with culture, not against it

The educational pipeline is not the only factor creating an imbalance in the numbers of women in technology. It’s a cultural challenge – in the US and abroad – too.

In India, we didn’t at first set out to make our Metiabruz center mostly women. Instead, we learned this through conversations with women who wanted to work with us while upholding their culture. By creating a safe space for women in a conservative society, we found we could maximize not just their comfort, but their participation, too.

Engaging more women has become one of our biggest goals, and it only makes sense to make sure their workplace is welcoming to them and sensitive to their culture.

In other contexts, this may mean ensuring that women – often seen as the sole family caretaker – are offered childcare options or extended maternity leave. Both of these have been shown to increase women’s labor force participation, and wages. In many cases, it means ensuring that we build a culture where women’s presence is respected (and ultimately, expected).

Everything from leave policies to company traditions and working hours can have gender-sensitive aspects. This is especially true when operating in a field that has been male-dominated for so long. It’s up to us to take note and build welcoming environments.

Connect women to growth-ready industries

Globally, women are an underutilized segment of the workforce. At the same time, the digital and data services industry is well-positioned for growth, making the joining of the two particularly powerful.

By linking women with high-growth industries and possibilities, we believe that we can fast-track their engagement not just in the workforce, but the digital future as a whole.

It’s not enough to incorporate women in technology generally; we’ve found that they must also be connected to rapidly-growing industries with huge future potential, like artificial intelligence and machine learning.

It Just Makes Sense

We can’t talk about diversity as if it is something that should only be pursued for its own sake. Encouraging women’s participation in technology (and other homogenous fields) should not be – and is not for us – a token gesture. Instead, encouraging diversity will bring about sustainable social change and can create sustainable businesses.

This is untapped potential. It is an opportunity to strengthen companies, families, communities, and individuals around the world.

0

Blog

Working with external teams to gather, clean and enhance datasets can be very rewarding. It minimizes the amount of repetitive work needed in-house, and can often bring even better data than working with your own teams. As we’ve talked about in previous weeks, however, it does come with its challenges. Learn about knowledge-transfer best practices here, and the importance of communication here. This week, we’re offering a more technical tip: […]

Working with external teams to gather, clean and enhance datasets can be very rewarding. It minimizes the amount of repetitive work needed in-house, and can often bring even better data than working with your own teams. As we’ve talked about in previous weeks, however, it does come with its challenges.

Learn about knowledge-transfer best practices here, and the importance of communication here.

This week, we’re offering a more technical tip: engineer your way through.

External teams will likely be completing tasks that require human intelligence. That doesn’t mean that engineering doesn’t play a key role. Instead, human intelligence can be augmented through automation or pre-processing of data.

In fact, by isolating the parts of the task that require human judgement from those that can be automated in some way, you’ll likely get even better results than with only automation or only human judgement.

What this means in practice is the development of simple interventions that make tasks easier and quicker. You could start with implementing keyboard shortcuts and auto-complete functions. While simple, they lighten the burden for your external teams, and improve their quality. If you’re working on a task like categorization (be it of items, comments, or service tickets), you could pre-process the dataset and tag items with potential categories. Then, external teams need only fill in the gaps, and remedy errors.

Along those lines, it is good practice to separate a big task into smaller tasks, each with their own skill needs. For example, you can have one team focus on segmenting an image, and another team on categorizing it, and yet another on transcribing included text. Each team develops proficiency in one skill, reaping rewards in terms of time to complete tasks and accuracy.

Finally, some cases are ripe for the use of APIs like Google Translate or RSS feeds to eliminate repetitive processes for your workers.

ecommerce

For an ecommerce marketplace client, we moderate posted items in order to identify repeat listings of the same item. (Items are often re-posted against marketplace rules in order to attempt to game search functionality.) We ran the task as presented to us by the client, and then again with a series of engineered interventions.

Ultimately, we found that three interventions created the best results and output:

  • Including a Google translate API to translate foreign language item names into English,
  • Incorporating worker-requested keyboard shortcuts throughout the process, and
  • Implementing automated string-comparison to prioritize obvious duplicates

With these three interventions, we were able to increase agreement by over 25 per cent.

Overall, the path to a good relationship with your external teams means remembering that though they’re external, they’re still a part of your team! Imagining them as new colleagues, who need an introduction to your company, who want to communicate, and who benefit from automated steps, will get you well on your way to the data you need.

0

Blog

As virtual reality enters its “adolescence” (and perhaps the trough of disillusionment), it languishes in an awkward stage where much has been accomplished, but perhaps it’s fallen short of all the hype. Arguments can – and should! – be made for a bright future, but at iMerit we have our eyes on another kind of reality: augmented reality. In some ways, augmented reality (AR) has pulled ahead of VR in […]

As virtual reality enters its “adolescence” (and perhaps the trough of disillusionment), it languishes in an awkward stage where much has been accomplished, but perhaps it’s fallen short of all the hype. Arguments can – and should! – be made for a bright future, but at iMerit we have our eyes on another kind of reality: augmented reality.

In some ways, augmented reality (AR) has pulled ahead of VR in terms of hype, and many watch as it pulls ahead in terms of promise as well. Big players like Microsoft are betting big on the promise of AR, and speculations around applications and profitability are broad.

Though at iMerit we’re aware of the challenges that face AR – everything from hardware concerns to gathering datasets to train computer algorithms – we are still excited. Here are a few reasons why.

1. Totally new tech.

Earlier this year, Microsoft released more information on its Holographic Processing Unit (HPU), which was quick to ignite excitement in many circles. The HPU gathers data from accelerometers and camera systems to not only recognize gestures, but also create a map of the environment in order to ensure that virtual objects stay put. Microsoft developed the HPU after they were unable to find any off-the-shelf vision-processing chips able to perform at the needed level. Now, it is about 200 times faster than a pure software equivalent, and operates at only 50 per cent of its capacity, meaning Microsoft is planning for it to do a whole lot more.

2. Novel ways to interact with technology.

By now, navigating the internet using a mouse, keyboard, and browser is second-nature to many of us. In the same way, we’ve quickly become adept at scrolling, tapping, and pinching our way through mobile apps and browsers. But what if there were even more ways to interact with our devices and technology? Until recently, progress on gesture tracking – especially of hands – has been slow. Understanding complex and subtle movement requires a lot of data and research. However, use of your hands is critical in an AR world, and increasing amounts of research open interesting possibilities that will ripple beyond the world of AR.

3. More dynamic and connected software

Good AR will collect huge amounts of information from the “real world” in order to augment it in a believable and exciting manner. The different inputs could take different forms – space maps, temperature, light levels – and they’ll need to be communicating efficiently and in real-time. This means that whatever software handling all of this input will need to be well-connected and powerful. The implications for this kind of software are huge, and could change not just AR, but the Internet of Things and more.

4. New jobs and markets

Given the rapid clip with which technology changes the field of job prospects, we should expect no less from VR and AR. Whether it’s creating entirely new positions – what does a VR Location Scout look like? – or repurposing old positions, the job possibilities are many. Udacity, a provider of online classes and nanodegrees, is now offering a new VR nanodegree that would prepare students for jobs in the emerging industry of VR. At iMerit, we work with clients building promising VR and AR systems, and employ hundreds of youth and women that would otherwise depend on informal, unstable jobs.

5. Applications as big as our imaginations

We might be getting used to some of the sillier (but still impressive) applications of augmented reality – see Snap or Facebook’s masks – but these are just the beginning. New use cases, both in-reach and still a ways off, abound. From letting fashionistas test out fashion week looks, to revolutionizing the way you take notes or do work, to changing the advertising and sportsindustries, you can bet that AR will touch thousands of companies around the world. The human-centric applications are just as ripe. Whether it’s making learning more engaging by bringing together VR/AR companies and education companies, or re-shaping the way healthcare is taught and practiced, we’re excited to see the ways AR can do good for our world.

0

Blog

When we think of applications for big data, we mostly imagine leveraging it for technology, analysis, and business insights. But what about sports? For the last year, iMerit has been working with KinaTrax to build accurate 3D models of all Major League Baseball (MLB) pitchers. These models are based on massive amounts of video data captured in-game from 12 different camera angles. iMerit teams sift through the video data and […]

Paving the Way for Big Data and Innovation in Sports

When we think of applications for big data, we mostly imagine leveraging it for technology, analysis, and business insights. But what about sports?

For the last year, iMerit has been working with KinaTrax to build accurate 3D models of all Major League Baseball (MLB) pitchers. These models are based on massive amounts of video data captured in-game from 12 different camera angles. iMerit teams sift through the video data and annotate pitchers’ movements throughout their pitch. These models are used by teams to monitor pitchers’ performance and to prevent injury. One team using this technology is the Chicago Cubs, who are headed to the World Series for the first time in 45 years.

Could this kind of technology be giving them the edge they need? Read more from Sport Techie here.

Update

Cubs win! Take a look at how iMerit and KinaTrax technology leveraged big data and contributed to this historic win. Read more.

0

PREVIOUS POSTSPage 3 of 9NEXT POSTS