Machine Archives - Business Intelligence Info

Tag Archives: Machine

Google details how it’s using AI and machine learning to improve search

October 16, 2020 Big Data

During a livestreamed event this afternoon, Google detailed the ways it’s applying AI and machine learning to improve the Google Search experience.

Soon, Google says users will be able to see how busy places are in Google Maps without having to search for specific beaches, parks, grocery stores, gas stations, laundromats, pharmacies, or other business, an expansion of Google’s existing busyness metrics. The company also says it’s adding COVID-19 safety information to business profiles across Search and Maps, revealing whether they’re using safety precautions like temperature checks, plexiglass, and more.

An algorithmic improvement to “Did you mean,” Google’s spell-checking feature for Search, will enable more accurate and precise spelling suggestions. Google says the new underlying language model contains 680 million parameters — the variables that determine each prediction — and runs in less than three milliseconds. “This single change makes a greater improvement to spelling than all of our improvements over the last five years,” Prabhakar Raghavan, head of Search at Google, said in a blog post.

Beyond this, Google says it can now index individual passages from webpages as opposed to whole pages. When this rolls out fully, it will improve roughly 7% of search queries across all languages, the company claims. A complementary AI component will help Search capture the nuances of what webpages are about, ostensibly leading to a wider range of results for search queries.

“We’ve applied neural nets to understand subtopics around an interest, which helps deliver a greater diversity of content when you search for something broad,” Raghavan continued. “As an example, if you search for ‘home exercise equipment,’ we can now understand relevant subtopics, such as budget equipment, premium picks, or small space ideas, and show a wider range of content for you on the search results page.”

Google is also bringing Data Commons, its open knowledge repository that combines data from public datasets (e.g., COVID-19 stats from the U.S. Centers for Disease Control and Prevention) using mapped common entities, to search results on the web and mobile. In the near future, users will be able to search for topics like “employment in Chicago” on Search to see information in context.

On the ecommerce and shopping front, Google says it has built cloud streaming technology that enables users to see products in augmented reality (AR). With cars from Volvo, Porsche, and other “top” auto brands, for example, they can zoom in to view the steering wheel and other details in a driveway, to scale, on their smartphones. Separately, Google Lens on the Google app or Chrome on Android (and soon iOS) will let shoppers discover similar products by tapping on elements like vintage denim, ruffle sleeves, and more.

Google details how it’s using AI and machine learning to improve search

Above: Augmented reality previews in Google Search.

Image Credit: Google

In another addition to Search, Google says it will deploy a feature that highlights notable points in videos — for example, a screenshot comparing different products or a key step in a recipe. (Google expects 10% of searches will use this technology by the end of 2020.) And Live View in Maps, a tool that taps AR to provide turn-by-turn walking directions, will enable users to quickly see information about restaurants including how busy they tend to get and their star ratings.

Lastly, Google says it will let users search for songs by simply humming or whistling melodies, initially in English on iOS and in more than 20 languages on Android. You will able to launch the feature by opening the latest version of the Google app or Search widget, tapping the mic icon, and saying “What’s this song?” or selecting the “Search a song” button, followed by at least 10 to 15 seconds of humming or whistling.

“After you’re finished humming, our machine learning algorithm helps identify potential song matches,” Google wrote in a blog post. “We’ll show you the most likely options based on the tune. Then you can select the best match and explore information on the song and artist, view any accompanying music videos or listen to the song on your favorite music app, find the lyrics, read analysis and even check out other recordings of the song when available.”

Google says that melodies hummed into Search are transformed by machine learning algorithms into a number-based sequence representing the song’s melody. The models are trained to identify songs based on a variety of sources, including humans singing, whistling, or humming, as well as studio recordings. They also take away all the other details, like accompanying instruments and the voice’s timbre and tone. This leaves a fingerprint that Google compares with thousands of songs from around the world and identify potential matches in real time, much like the Pixel’s Now Playing feature.

“From new technologies to new opportunities, I’m really excited about the future of search and all of the ways that it can help us make sense of the world,” Raghavan said.

Last month, Google announced it will begin showing quick facts related to photos in Google Images, enabled by AI. Starting in the U.S. in English, users who search for images on mobile might see information from Google’s Knowledge Graph — Google’s database of billions of facts — including people, places, or things germane to specific pictures.

Google also recently revealed it’s using AI and machine learning techniques to more quickly detect breaking news around crises like natural disasters. In a related development, Google said it launched an update using language models to improve the matching between news stories and available fact checks.

In 2019, Google peeled back the curtains on its efforts to solve query ambiguities with a technique called Bidirectional Encoder Representations from Transformers, or BERT for short. BERT, which emerged from the tech giant’s research on Transformers, forces models to consider the context of a word by looking at the words that come before and after it. According to Google, BERT helped Google Search better understand 10% of queries in the U.S. in English — particularly longer, more conversational searches where prepositions like “for” and “to” matter a lot to the meaning.

BERT is now used in every English search, Google says, and it’s deployed across languages including Spanish, Portuguese, Hindi, Arabic, and German.

Let’s block ads! (Why?)

Big Data – VentureBeat

The secrets of small data: How machine learning finally reached the enterprise

October 9, 2020 Big Data

Over the past decade, “big data” has become Silicon Valley’s biggest buzzword. When they’re trained on mind-numbingly large data sets, machine learning (ML) models can develop a deep understanding of a given domain, leading to breakthroughs for top tech companies. Google, for instance, fine-tunes its ranking algorithms by tracking and analyzing more than one trillion search queries each year. It turns out that the Solomonic power to answer all questions from all comers can be brute-forced with sufficient data.

But there’s a catch: Most companies are limited to “small” data; in many cases, they possess only a few dozen examples of the processes they want to automate using ML. If you’re trying to build a robust ML system for enterprise customers, you have to develop new techniques to overcome that dearth of data.

Two techniques in particular — transfer learning and collective learning — have proven critical in transforming small data into big data, allowing average-sized companies to benefit from ML use cases that were once reserved only for Big Tech. And because just 15% of companies have deployed AI or ML already, there is a massive opportunity for these techniques to transform the business world.

The secrets of small data: How machine learning finally reached the enterprise

Above: Using the data from just one company, even modern machine learning models are only about 30% accurate. But thanks to collective learning and transfer learning, Moveworks can determine the intent of employees’ IT support requests with over 90% precision.

Image Credit: Moveworks

From DIY to open source

Of course, data isn’t the only prerequisite for a world-class machine learning model — there’s also the small matter of building that model in the first place. Given the short supply of machine learning engineers, hiring a team of experts to architect an ML system from scratch is simply not an option for most organizations. This disparity helps explain why a well-resourced tech company like Google benefits disproportionately from ML.

But over the past several years, a number of open source ML models — including the famous BERT model for understanding language, which Google released in 2018 — have started to change the game. The complexity of creating a model the caliber of BERT, whose aptly named “large” version has about 340 million parameters, means that few organizations can even consider quarterbacking such an initiative. However, because it’s open source, companies can now tweak that publicly available playbook to tackle their specific use cases.

To understand what these use cases might look like, consider a company like Medallia, a Moveworks customer. On its own, Medallia doesn’t possess enough data to build and train an effective ML system for an internal use case, like IT support. Yet its small data does contain a treasure trove of insights waiting for ML to unlock them. And by leveraging new techniques to glean these insights, Medallia has become more efficient, from recognizing which internal workflows need attention to understanding the company-specific language its employees use when asking for tech support.

Massive progress with small data

So here’s the trillion-dollar question: How do you take an open source ML model designed to solve a particular problem and apply that model to a disparate problem in the enterprise? The answer starts with transfer learning, which, unsurprisingly, entails transferring knowledge gained from one domain to a different domain that has less data.

For example, by taking an open source ML model like BERT — designed to understand generic language — and refining it at the margins, it is now possible for ML to understand the unique language employees use to describe IT issues. And language is just the beginning, since we’ve only begun to realize the enormous potential of small data.

Above: Transfer learning leverages knowledge from a related domain — typically one with a greater supply of training data — to augment the small data of a given ML use case.

Image Credit: Moveworks

More generally, this practice of feeding an ML model a very small and very specific selection of training data is called “few-shot learning,” a term that’s quickly become one of the new big buzzwords in the ML community. Some of the most powerful ML models ever created — such as the landmark GPT-3 model and its 175 billion parameters, which is orders of magnitude more than BERT — have demonstrated an unprecedented knack for learning novel tasks with just a handful of examples as training.

Taking essentially the entire internet as its “tangential domain,” GPT-3 quickly becomes proficient at these novel tasks by building on a powerful foundation of knowledge, in the same way Albert Einstein wouldn’t need much practice to become a master at checkers. And although GPT-3 is not open source, applying similar few-shot learning techniques will enable new ML use cases in the enterprise — ones for which training data is almost nonexistent.

The power of the collective

With transfer learning and few-shot learning on top of powerful open source models, ordinary businesses can finally buy tickets to the arena of machine learning. But while training ML with transfer learning takes several orders of magnitude less data, achieving robust performance requires going a step further.

That step is collective learning, which comes into play when many individual companies want to automate the same use case. Whereas each company is limited to small data, third-party AI solutions can use collective learning to consolidate those small data sets, creating a large enough corpus for sophisticated ML. In the case of language understanding, this means abstracting sentences that are specific to one company to uncover underlying structures:

Above: Collective learning involves abstracting data — in this case, sentences — with ML to uncover universal patterns and structures.

Image Credit: Moveworks

The combination of transfer learning and collective learning, among other techniques, is quickly redrawing the limits of enterprise ML. For example, pooling together multiple customers’ data can significantly improve the accuracy of models designed to understand the way their employees communicate. Well beyond understanding language, of course, we’re witnessing the emergence of a new kind of workplace — one powered by machine learning on small data.

Let’s block ads! (Why?)

Big Data – VentureBeat

Measuring & Optimizing Overall Equipment Effectiveness with Machine Learning

October 7, 2020 BI News and Info

Overall equipment effectiveness (OEE) is a metric used in manufacturing operations to see and understand how efficiently processes and equipment are being used. It looks at facilities, time, materials, and the productivity of each of those discrete aspects of the manufacturing process. Given the budget constraints that all manufacturers face, it’s obvious that equipment effectiveness measures are so popular and necessary.

To dive into the concept of overall equipment effectiveness further, there are three major aspects that you must consider when it comes to setting key performance indicators (KPIs) for OEE.

Availability: The amount of time that the equipment is ready and useful for operation
Performance: How well the equipment works at maximum operating speed
Quality: The percentage of ‘good’ parts that are produced (as opposed to parts that need to be scrapped)

Key Performance Indicators for Overall Equipment Effectiveness

Because each of these KPIs is quantifiable, there are calculations managers can use to benchmark their performance, thus allowing them to find places for improvement and then dial up the optimal outputs for their processes. The KPIs are fairly straightforward to calculate.

Availability is measured as the portion of time that a given machine is ready to operate. It is sometimes referred to as ‘uptime’, and is designated as a simple ratio of operating time over scheduled time, where operating time takes into account the actual measured time that the machine is used, and scheduled time would be, for example, the time that the machine is used on a shift.

Performance is measured as a ratio of the actual pace of the machine over the designed speed of the machine, and as such doesn’t consider either quality or availability. As an uncomplicated example, if a machine can travel at 100 MPH, and it’s only driven at 50 MPH, the performance would be shown as 50%.

Quality is measured as the ratio of good parts to the overall total of the parts output of the machine. This is also considered the yield.

Key Benefits of Using Overall Equipment Effectiveness

The old saying ‘you can’t manage what you can’t measure’ applies quite well to OEE. Having the ability to measure each of these KPIs means manufacturers that adopt OEE can better manage their processes.

Some of the benefits of using OEE as a tool for manufacturing include:

Optimizing machine usage: Developing an understanding of a machine’s performance allows manufacturers to optimize that performance with subtle adjustments
Improving process quality: Producing fewer defective products means less waste and better ROI
Reducing repair costs: Knowing the expected machine efficiency means that proactive measures can be taken to repair prior to major breakdowns

Optimizing Overall Equipment Effectiveness with ML

From an OEE perspective, it’s a best practice to measure at the step in the process where there’s a bottleneck, or potential constraint. No matter what’s being manufactured, there’s always a point in the process that can become an obstacle. It’s at that point that OEE is critical to understand what’s happening, as it’s there that determines the overall performance.

In the past, measuring OEE and making adjustments based on those measures was something that happened largely manually, and was based on historical knowledge. But it shouldn’t come as any surprise that each of these is an ideal use case for machine learning. As we describe in our blog post on predictive maintenance, the major driver for the Industry 4.0 Revolution is the rapid development of the Internet of Things (IoT).

As sensors become more embedded in machines, as well as integral to manufacturing, it becomes easier for manufacturers to automatically measure the necessary components to optimize their intricate operations. The data that is generated by the sensors can be used to ensure that the appropriate machine learning algorithms have enough data to be useful.

For the above-mentioned three major aspects of OEE (availability; performance; quality), here’s how machine learning can be leveraged:

Availability: Machine learning algorithms can help lower the amount of time needed to setup or retool manufacturing lines, based on previous similar occurrences, helping to increase OEE
Performance: The data gathered can help an ML algorithm identify roadblocks or slowdowns in production, and then leverage predictive maintenance to lessen or eliminate them
Quality: ML algorithms can be applied to increase the usable manufacturing yields of a process

Final Thoughts

OEE is a valuable tool in almost every manufacturing operation and, by using the proper machine learning techniques, manufacturers can truly optimize their operational efficiencies, easily and automatically.

If you’re curious to learn more about the effects that AI and ML are having on your manufacturing processes, sign up for a free, no obligation AI assessment. We’ll help you explore the most impactful ways you can use AI in your operations.

Let’s block ads! (Why?)

RapidMiner

10 machine learning algorithms you need to know

September 7, 2020 BI News and Info

If you’ve just started to explore the ways that machine learning can impact your business, the first questions you’re likely to come across are what are all of the different types of machine learning algorithms, what are they good for, and which one should I choose for my project? This post will help you answer those questions.

There are a few different ways to categorize machine learning algorithms. One way is based on what the training data looks like. There are three different categories used by data scientists with respect to training data:

Supervised, where the algorithms are trained based on labeled historical data—which has often been annotated by humans—to try and predict future results.
Unsupervised, by contrast, uses unlabeled data that the algorithms try to make sense of by extracting rules or patterns on their own.
Semi-supervised, which is a mix of the two above methods, usually with the preponderance of data being unlabeled, and a small amount of supervised (labeled) data.

Another way to classify algorithms—and one that’s more practical from a business perspective—is to categorize them based on how they work and what kinds of problems they can solve, which is what we’ll do here.

There are three basic categories here as well: regression, clustering, and classification algorithms. Let’s jump into each.

Regression algorithms

There are basically two kinds of regression algorithms that we commonly see in business environments. These are based on the same regression that might be familiar to you from statistics.

1. Linear regression

Described very simply, linear regression plots a line based on a set of data points, called the dependent variable (plotted on the y-axis) and the explanatory variable (plotted on the x-axis).

Linear regression is a commonly used statistical model that can be thought of as a kind of Swiss Army knife for understanding numerical data. For example, linear regression can be used to understand the impact of price changes on goods and services by mapping the sales of various prices against its sales, in order to help guide pricing decisions. Depending on the specific use case, some of the variants of linear regression, including ridge regression, lasso regression, and polynomial regression might be suitable as well.

2. ARIMA

ARIMA (“autoregressive integrated moving average”) models can be considered a special type of regression model.

It allows you to explore time-dependent data points because it understands data points as a sequence, rather than as independent from one another. For this reason, ARIMA models are especially useful for conducting time-series analyses, for example, demand and price forecasting.

Clustering algorithms

Clustering algorithms are typically used to find groups in a dataset, and there’s a few different types of algorithms that can do this.

3. k-means clustering

k-means clustering is generally used to segregate groups with related characteristics and group them together.

Businesses looking to develop customer segmentation strategies might use k-means clustering to better target marketing campaigns that groups of customers should respond to. Another use case for k-means clustering would be detecting insurance fraud, using historical data that in the past had showed tendencies to defraud the insurance provider to examine current cases.

4. Agglomerative & divisive clustering

Agglomerative clustering is a method used for finding hierarchal relationships for data clusters.

It uses a bottom-up approach, putting each individual data point into its own cluster, and then merging similar clusters together. By contrast, divisive clustering takes the opposite approach, and assumes all the data points are in the same cluster and then divides similar clusters from there.

A timely use case for these clustering algorithms is tracking viruses. By using DNA analysis, scientists are able to better understand mutation rates and transmission patterns.

Classification algorithms

Classification algorithms are similar to clustering algorithms, but while clustering algorithms are used to both find the categories in data and sort data points into those categories, classification algorithms sort data into predefined categories.

5. k-nearest neighbors

Not to be confused with k-means clustering, k-nearest neighbors is a pattern classification method that looks at the data presented, scans through all past experiences, and identifies the one that is the most similar.

k-nearest neighbors is often used for activity analysis in credit card transactions, comparing transactions to previous ones. Abnormal behavior, like using a credit to make a purchase in another country, might trigger a call from the card issuers fraud detection unit. The algorithm can also be used for visual pattern recognition, and it’s now frequently used as part of retailers’ loss prevention tactics.

6. Tree-based algorithms

Tree-based algorithms, including decision trees, random forests, and gradient-boosted trees are used to solve classification problems. Decision trees excel at understanding data sets that have many categorical variables and can be effective even when some data is missing.

They’re primarily used for predictive modeling, and are helpful in marketing, answering questions like “which tactics should we be doing more of?” A decision tree might help an email marketer decide which customers would be more likely to order based on specific offers.

A random forest algorithm uses multiple trees to come up with a more complete analysis. In a random forest algorithm, multiple trees are created, and the forest uses the average decisions of its trees to make a prediction.

Gradient-boosted trees (GBTs) also use decision trees but rely on an iterative approach to correct for any mistakes in the individual decision tree models. GBTs are widely considered to be one of the most powerful predictive methods available to data scientists and can be used by manufacturers to optimize the pricing of a product or service for maximum profit, among other use cases.

7. Support vector machine

A support vector machine (SVM) is, according to some practitioners, the most popular machine learning algorithm. It’s a classification (or sometimes a regression) algorithm that’s used to separate a dataset into classes, for example two different classes might be separated by a line that demarcates a distinction between the classes.

There could be an infinite number of lines that do the job, but SVM helps find the optimal line. Data scientists are using SVMs in a wide variety of business applications, including classifying images, detecting faces, recognizing handwriting, and bioinformatics.

8. Neural networks

Neural networks are a set of algorithms designed to recognize patterns and mimic, as much as possible, the human brain. Neural nets, like the brain, are able to adapt to changing conditions, even ones that weren’t originally intended.

A neural net can be taught to recognize, say, an image of a dog by providing a training set of images of dogs. Once the algorithm processes the training set, it can then classify novel images into ‘dogs’ or ‘not dogs’. Neural networks work on more than just images, though, and can be used for text, audio, time-series data, and more. There are many different types of neural networks, all optimized for the specific tasks they’re intended to work on.

Some of the business applications for neural networks are weather prediction, face detection and recognition, transcribing speech into text, and stock market forecasting. Marketers are using neural networks to target specific content and offers to customers who would be most ready to act on the content.

Deep learning is really a subset of neural networks, where algorithms ‘learn’ by analyzing large datasets. Deep learning has a myriad of business uses, and in many cases, it can outperform the more general machine learning algorithms. Deep learning doesn’t generally require human inputs for feature creation, for example, so it’s good at understanding text, voice and image recognition, autonomous driving, and many other uses.

Other algorithm types

In addition to the above categories, there are other types of algorithms that can be used during model creation and training to help the process, like fuzzy matching and feature selection algorithms.

9. Fuzzy matching

Fuzzy matching is a type of clustering algorithm that can make matches even when items aren’t exactly the same, due to data issues like typos. For some natural language processing tasks, preprocessing with fuzzy matching can improve results by three to five percent.

A typical use case is customer profile management. Fuzzy matching lets you identify very similar addresses as the same so that only one unique record ID and source file would be used for the two similar addresses.

10. Feature selection algorithms

Feature selection algorithms are used to whittle down the number of input parameters from a model. A lower number of input variables may lower the computational cost of running a model, as well as improve the performance of the model.

The commonly used techniques like PCA and MRMR are useful for picking up as much information as possible from a reduced subset of features. Using a subset of features can be beneficial because your model may be less confused by noise and the computation time of your algorithm will go down. Feature selection has been used to show business competitor relationships, for example.

If you want to dive deeper into machine learning, including how to get your first project off the ground, check out RapidMiner’s Human’s Guide to Machine Learning Projects.

Let’s block ads! (Why?)

RapidMiner

Teradata is Selected by Brinker International to Enhance Advanced Analytics, Machine Learning and Data Science Capabilities

August 21, 2020 BI News and Info

Leading Casual Dining Restaurant Company Reinvests in Teradata as it Moves from On-Premises to the Cloud

Teradata (NYSE: TDC), the cloud data and analytics company, today announced that after an evaluation of other cloud analytics offerings on the market, Brinker International, Inc. (NYSE: EAT) has reinvested with Teradata, leveraging the Teradata Vantage platform – delivered as-a-service, on Amazon Web Services (AWS) – as the core of its data foundation to facilitate advanced analytics, machine learning and data science across the organization.

Brinker is one of the world’s leading casual dining restaurant companies and has been a Teradata customer for more than two decades. Founded in 1975 and based in Dallas, Texas, Brinker owns, operates, or franchises more than 1,600 restaurants under the names Chili’s® Grill & Bar and Maggiano’s Little Italy®. Over the past year, Brinker has been working to further increase its capabilities in advanced analytics and data science.

“Being a data-driven organization allows us to make informed decisions to create a better Guest and Team Member experience,” said Pankaj Patra, senior vice president and chief information officer at Brinker International. “As we looked for more flexible and cost-effective ways to manage and access our data, we evaluated quite a few cloud-native providers. After careful consideration, we decided the best course of action would be to migrate to Teradata Vantage in the cloud and take advantage of its as-a-service offerings to support our analytic goals.”

With Teradata Vantage delivered as-a-service, in the cloud, enterprises such as Brinker can focus on mining their data for insights that drive business decisions, rather than on managing infrastructure. By integrating Vantage’s machine learning capabilities, Brinker can now apply advanced analytics and predictive modeling to its business processes, enabling more accurate sales forecasting, demand and traffic forecasting, team member management, recommendation engines for customers and more.

“We’re proud of our ongoing relationship with Brinker and its long-standing position as a leader in the restaurant industry – a position due in large part to its culture of innovation in using data and analytics to streamline business processes, facilitate rapid decision-making and turn insights into answers,” said Ashish Yajnik, vice president of Vantage Cloud at Teradata. “Our collaboration with AWS and participation in the AWS Independent Software Vendor (ISV) Workload Migration Program has helped Brinker successfully move their mission-critical data infrastructure to the cloud. We look forward to expanding our relationship by powering their advanced analytics and data science capabilities through the scalable, clean and trusted data foundation that the Vantage platform provides.”

Teradata is an Advanced Technology and Consulting Partner in the AWS Partner Network (APN). The company brings proven processes and tools to make migrations to Vantage on AWS low risk and the fastest path to customer value through the AWS ISV Workload Migration – an APN Partner program that helps customers migrate ISV workloads to AWS to achieve their business goals and accelerate their cloud journey.

“Through the AWS ISV Workload Migration Program, Teradata was able to help Brinker migrate to Vantage on AWS securely and cost effectively. We are pleased to collaborate with Teradata and its long-standing customer Brinker to enhance their cloud practices,” said Sabina Joseph, director, Americas ISVs, Amazon Web Services, Inc.

Teradata Vantage is the leading hybrid cloud data analytics software platform that enables ecosystem simplification by unifying analytics, data lakes and data warehouses. With Vantage delivered as-a-service, enterprise-scale companies can eliminate silos and cost-effectively query all their data, all the time, regardless of where the data resides – in the cloud using low cost object stores, on multiple clouds, on-premises or anywhere in-between – to get a complete view of their business. And by combining Vantage with first party cloud services, Teradata enables customers to expand their cloud ecosystem with deep integration of cloud-specific, cloud-native services.

Webinar
Join Teradata for a live webinar on July 29th, 8:00 – 9:00 a.m. PT featuring Mark Abramson, lead architect, BI and analytics at Brinker International, and William McKnight, president of McKnight Consulting Group. The session will be moderated by Ed White, vice president, portfolio marketing and competitive intelligence at Teradata. Details below:

Webinar: Brinker’s Journey Back to Teradata

Wednesday, July 29th
8:00 a.m. – 9:00 a.m. PT /
11:00 a.m. – 12:00 p.m. ET

Registration is required and is open to Teradata prospects, customers, analysts, partners and Teradata employees.

This interactive webinar will highlight:

Brinker’s future analytic strategies and how Teradata will be part of its ongoing journey to lower overall costs and improve performance.
How Brinker embraces and drives benefits using Teradata Vantage on AWS, particularly to meet their advanced analytics and computing needs.
McKnight’s latest research into price-performance on modern cloud database management systems, including best practices.

About Brinker International, Inc.
Hi, welcome to Brinker International, Inc. (NYSE: EAT)! We’re one of the world’s leading casual dining restaurant companies. Founded in 1975 in Dallas, Texas, we stay true to our roots, but also enjoy exploring outside of our hometown. As of March 25, 2020, we owned, operated or franchised 1,675 restaurants in 29 countries and two territories under the names Chili’s® Grill & Bar (1,622 restaurants) and Maggiano’s Little Italy® (53 restaurants). Our passion is making people feel special and we hope you feel that passion each time you visit one of our restaurants or our home office. Find more information about us at www.brinker.com, follow us on LinkedIn or review us on Glassdoor.

Let’s block ads! (Why?)

Teradata United States

Machine learning groups form Consortium for Python Data API Standards to reduce fragmentation

August 18, 2020 Big Data

Deep learning framework Apache MXNet and Open Neural Network Exchange (ONNX) today launched the Consortium for Python Data API Standards, a group that wants to make it easier for machine learning practitioners and data scientists no matter which framework, library, or tool from the Python ecosystem it came from. ONNX is a group initially formed by Facebook and Microsoft in 2017 to power interoperability between frameworks and tools. Today the group includes near 40 organizations influential in AI and data science like AWS, Baidu, and IBM as well as hardware makers like Arm, Intel, and Qualcomm.

The group, which will develop standards for dataframes and arrays or tensors, said the consortium is necessary due to fragmentation of the kinds of frameworks of the data ecosystem in recent years.

Other major frameworks include TensorFlow, PyTorch, and NumPy; the Python programming language is also used for Python dataframes like Pandas, PySpark, and Apache Arrow. PyTorch, one of the most popular machine learning frameworks in use today is not a part of the consortium, a Facebook company spokesperson told VentureBeat in an interview.

“Currently, array and dataframe libraries all have similar APIs, but with enough differences that using them interchangeably isn’t really possible,” group members said in a blog post today. “We aim to grow this Consortium into an organization where cross-project and cross-ecosystem alignment on APIs, data exchange mechanisms and other such topics happens. These topics require coordination and communication to a much larger extent than they require technical innovation. We aim to facilitate the former, while leaving the innovating to current and future individual libraries.”

Initial efforts will start with a working group then request feedback from array and dataframe library maintainers and iterate before the first version of the standard is made available for use. The first feedback session begins next month. As part of the launch, the group is releasing tools for comparing array or tensor and tracking some of the primary functions of a dataframe library.

While AI research dates back to the 1950s, the practical need to create standards and build an infrastructure for benchmark testing, interoperability, and other practical developer needs led to the formation of groups like ONNX. Beyond machine learning, other examples of tech groups formed to create standards include C++ and Open Geospatial Consortium.

Let’s block ads! (Why?)

Big Data – VentureBeat

The 5 Stages of Machine Learning and the Unique Data Requirements of Each

July 29, 2020 TIBCO Spotfire

Reading Time: 2 minutes

The importance of data management in the successful deployment of data science and machine learning (ML) cannot be overemphasized. Of course, data is that elephant in the room that no one wants to talk about. It’s that gnarly beast that keeps data scientists, citizen data scientists, and data analysts up at night. But the integrity of the data that you use in processing your advanced analytics is crucial to achieving the results that every business is looking for: those pivotal insights that could potentially be game-changing.

In a new best practices report, Data Management for Advanced Analytics, 2020, research firm Transforming Data with Intelligence helps shed some light on how to get data management right for one of the hottest trends in advanced analytics today–machine learning.

According to the report, “Machine Learning and AutoML are only as good as the data fed to them. Here’s the catch: the newfound abilities of machine learning and AutoML depend heavily on getting the right data at the right time to the correct models” (TDWI, Data Management for Advanced Analytics, 2020, Philip Russom).

Machine Learning and AutoML are only as good as the data fed to them. Click To Tweet

In the report, Russom lays out the five cycles of machine learning and the different, yet equally demanding data requirements for each cycle. The cycles that Russom alludes to in machine learning are:

Solution definition
Development
Deployment
Production
Monitoring output

To make the content more easily digestible and understandable, Russom provides a diagram to illustrate the different data requirements by stage:

The 5 Stages of Machine Learning and the Unique Data Requirements of Each — *Source: TDWI 2019.1*

As you can see, ensuring that machine learning models have the right data at the right time is a complex task. Use this report to make sure you satisfy all the requirements of each stage of machine learning – the future of your business may depend on it.

Let’s block ads! (Why?)

The TIBCO Blog

Microsoft and TIBCO Team Up to Bring Machine Learning and AI to IoT and the Edge

July 18, 2020 TIBCO Spotfire

Reading Time: 2 minutes

In our continuing commitment to accelerate digital business transformation through the use of artificial intelligence (AI) and machine learning (ML), TIBCO unveiled the capabilities of TIBCO Spotfire^® and TIBCO^®Data Science to support Microsoft Azure Cognitive Services at a recent Build conference. Thanks to this combo of TIBCO and Microsoft technologies, sensor data, and log data can now be analyzed on edge devices with little to no internet connectivity. This means less lag time for analytics and hence, faster response times for businesses based on that intelligence, particularly in places with unreliable internet connections.

The need for intelligence at the edge without internet connectivity is growing fast. In fact, in a report by Gartner, they predict that “By 2022, more than 50% of enterprise-generated data will be created and processed outside the data center or cloud.” As more devices become intelligent, there will be a greater need for processing the data they collect outside of core systems and closer to the edge.

Think of factory floors, shipping containers at sea, and any other location where the internet is spotty. Now, in places with little to no internet connectivity, developers can bring the power of TIBCO visual analytics and data science solutions to detect and prevent anomalies. Anomalies help improve their organization’s operational efficiency and lower equipment costs.

“Increasingly, customers performing predictive maintenance must look for ways to leverage AI closer to where their data is generated for timely analytics and responses,” said Matt Quinn, Chief Operating Officer, TIBCO. So, for those who have focused on bringing AI to their core systems, it’s now time to think about how you can extend intelligence to your Internet of Things (IoT) and edge devices that might lack full-time internet connectivity.

For those who have focused on bringing AI to their core systems, it’s now time to think about how you can extend intelligence to your Internet of Things (IoT) and edge devices that might lack full-time internet connectivity. Click To Tweet

Some excellent use cases highlighting the power of intelligence at the edge where internet connectivity is questionable or shaky have been popping up in recent years including the University of Iowa Hospitals and Clinics who are proactively predicting infection rates using advanced analytics on data collected on medical devices. Their efforts saw a drop of 74 percent in onsite surgical infections as a result of making predictive decisions directly in operating rooms. Anadarko, an oil and gas exploration and production company that has been working with TIBCO for several years, increased the speed of their drilling projects by 10 times even though many of their drill sites are in the middle of the ocean or in underdeveloped countries where internet connectivity is spotty at best.

With so much data being generated at the edge, it’s not always practical to bring that data back to a central location for processing. You need the capabilities of AI and ML to bring better and faster decisioning to the edge without the limitations of, slow or no internet connectivity. For more information on how the TIBCO plus Microsoft use case might be able to help your company realize machine learning at the edge, please visit Microsoft Channel 9 for the demo given at the Build Conference, read the TechTarget News article or TIBCO press release for more details.

Let’s block ads! (Why?)

The TIBCO Blog

10 surprising applications of machine learning in everyday life

July 10, 2020 BI News and Info

There’s no doubt about it—machine learning and artificial intelligence have significantly changed our world over the last few decades.

Every day there seems to be a new breakthrough about how AI and ML can be leveraged for positive development, to the point that it’s sometimes hard to tell what’s real and what’s being sensationalized.

With headlines about AI replicating our brains or reading our minds or driving our cars, we often miss the many simple yet genius applications of machine learning that are right in front of us.

Applications of Machine Learning in Everyday Life

To help bring us back down to Earth, we wanted to look at some of the surprising and interesting ways that ML is impacting our daily lives, whether we see it or not.

1. Managing traffic

AI excels at managing logistics, and managing a city’s traffic is one of the most complicated and difficult pieces of logistics to understand. Millions of datapoints and thousands of commuters must be properly accounted for, in effort to ensure the trains run on time, traffic lights are functioning correctly, and maintenance is performed where and when it’s needed.

In many modern cities, machine learning needs to be at the center of the transportation system. Data like rush hour fluctuations, road wear and tear, and existing maintenance schedules can be used to build powerful ML models. Ultimately, these models help improve traffic flow, increase the usage and efficiency of sustainable modes of transportation, and limit real-world disruption by modelling and visualizing future changes.

2. Making beer

Despite traditionally involving only three ingredients (water, barley, and hops), the art of brewing beer is a complicated and precise scientific process. The simplest change can dramatically affect the flavor and even ruin a batch—whether it’s the temperature of which it’s brewed or the amount of time the brew sits for. And that’s without considering the countless ingredients that are used in modern brewing techniques.

With machine learning algorithms, brewers can more easily make and keep track of these micro-adjustments to ensure flavor quality and consistency. ML can also be used to develop and refine new flavors by analyzing past popular and successful brews, as well as brews that failed or were less popular.

Breweries can even make batches specific to one customer’s tastes by tracking their past buying habits and beer preferences. Best of all, this technology is scalable and can be applied to as large or small a batch as the brewer wants.

3. Better Translation

How many times have you pulled out your phone to translate a phrase you don’t understand? Or searched how to properly pronounce a word with more vowels than you thought was possible? Speech and translation technology are nothing new, having been available via various tools for years. But AI and ML are making huge changes in just how effective the technology is.

Translating two related, modern languages like French and Spanish is one thing, but translating a 2,500-year-old, hand-carved dead language into modern English is a painstaking and time-consuming task even for expert archeologists.

Luckily, artificial intelligence-powered computer vision is making it possible to automate the translation of the ancient Persian cuneiform, giving researchers far more material to examine and helping to ensure accuracy. This technology doesn’t just work on ancient tablets though; it can be used on any language to translate even handwritten cursive into easy to read print.

ML isn’t just limited to translating either; researchers are working on technology that translates brain signals into speech with 97% accuracy. While mind-reading AI is still a ways off, this technology could soon help people without the ability to speak to communicate in ways they never could otherwise.

4. Creating video games

The video game industry, more than many others, is constantly pushing the technological standard in order to present the best experience for players. Developers are no strangers to using rudimentary AI algorithms in their games.

Now, AI is even creating games in real-time for players to interact with, with no underlying game engine needed. For Pac-Man’s 40^th anniversary, researchers at NVIDIA trained an AI on 50,000 playthroughs of the game, with the end result being a completely ML-powered edition of Pac-Man.

This technology goes well beyond recreating arcade classics, of course. Game developers can use the underlying technology to add details to environments in real time, create enemies that can dynamically learn from players’ behaviors, or create entire levels that change dynamically based on players’ actions.

5. Redefining music

Music has constantly been evolving ever since a caveman banged two sticks together, and technology has always been the biggest factor in pushing the art form’s boundaries.

Now, machine learning is posed to redefine just what music is. Researchers have developed AIVA (Artificial Intelligence Virtual Artist) and taught it how to compose music in a classical style.

AIVA was built using deep learning and reinforcement algorithms, allowing AI to develop an understanding of what makes music sound pleasing to human ears, while still being able to learn and improve its music over time. This also helps AIVA to insert some variety into its music, avoiding each composition sounding the same.

The most impressive thing about AIVA is that its music isn’t just theoretical. AIVA’s compositions have already began to appear in commercials, film, and video games, with the copyrights to each piece being under AIVA’s name.

6. Moderating social media

Proper moderation, filtering, and routing of messages has been a problem for a variety of businesses that deal with customer service issues, and have been an especially difficult problem for social media platforms, with no real solution proving to be viable—at least not until now.

With the current climate of misinformation around COVID-19 and the pervasive use of hate speech on online forums, Facebook has turned to artificial intelligence to help keep its platform in check. The AI they developed will act in a role like that of the human moderators, some of which it is replacing.

Facebook’s algorithm is trained to recognize hate speech and fact-check misinformation. Some of the key features it looks for are duplicate images, which often indicate a meme or infographic that’s circulated by bots or trolls on the platform. The difficulty with this solution is the risk that images and memes using the same format may get banned, even if they aren’t being used for nefarious purposes.

While researchers are hopeful that the AI will be able to detect when a post is similar but not the same as banned content, this is still a job for humans in many cases. Nevertheless, machine learning is making it much easier for moderators to better police Facebook and other social media platforms.

7. Deciding what to eat

Do you feel like Mexican food for lunch? Italian? Dine in or eat out? Luckily, machine learning can help you make those decisions and find exactly what you want.

Simple forms of this technology exist already in restaurant reviews apps, map apps, and food delivery services that track what kind of food you like to eat and make predictions about what you might want based on your location, history, and time of day. But Kartik Dhawan, writing on Towards Data Science, took it a step further by using text analytics to determine what kind of ramen noodles he wanted, down to the region, ingredients, and flavor profile.

Dhawan compiled data from a ramen reviews site and performed text analysis on the ratings and relative frequency of specific flavors and ingredients to determine which ‘themes’ were most popular among Ramen lovers. By combining this analysis with brand and region ratings, Dhawan was able to pick the best variety of ramen for himself. This technology has much broader applications in the future as well, with services analyzing your reviews, ratings, region, and dining history to determine the best dish for you.

8. Tutoring students

The current COVID-19 pandemic is forcing students and teachers to adapt to new ways of learning and teaching, especially while not in-person. Luckily, AI is proving an invaluable tool in this new teaching environment.

Researchers at Carnegie Mellon University developed an machine learning tutor that can dynamically learn from a teacher by attempting to solve problems on its own, and using corrections from its teacher to extrapolate solutions to other problems, as well as alternate methods for solving the problem that even the teacher may have not thought of.

The AI can then use the different techniques it’s learned to teach students in the same way and adapt to any difficulties that the students might have. For instance, if a student has difficulty in understanding the order of operations in a larger math problem, AI can respond dynamically to the student’s specific question in a way a video or virtual class would be unable to do effectively.

9. Helping to fight COVID-19

Just about every industry—from automotive manufacturers to liquor distilleries—have stepped up to help in light of the current crisis, and the data science community is no exception. Machine learning has been absolutely instrumental in tracking and projecting the spread of the virus, and now it’s helping slow the spread as well.

Workplaces and stores that are carefully reopening have integrated machine learning algorithms to help ensure employees and customers are as safe as possible. Existing security cameras are being equipped with AI to identify when people are not properly social distancing or wearing masks and informing managers accordingly.

While this technology already existed in other forms—for example, monitoring whether or not safety gear is being worn in factory settings—its broader adoption will likely lead to more safety and security developments, even after the epidemic ends.

10. Developing new fragrances

For perfumeries, the choice to integrate AI into their processes is a no-brainer. Perfumes are a delicate balance between art and chemistry, and machine learning is helping to combine the two seamlessly and inexpensively.

One fragrance startup, Scentbird, uses a custom-built ML model with data gathered from customers to develop specialized and unique scents for its human counterparts to test. Another company, Symrise, has its own dedicated AI perfumer, called Phylra. While Scentbird’s ML models generate scents for human approval, Phylra is free to make its own decisions and evaluate how customers respond to fragrances all on its own.

ML also helps identify where alternative ingredients can be used, driving the price of manufacturing down while using more natural, environmentally safe ingredients.

Wrapping Up

When it comes to machine learning, the possibilities are virtually limitless. ML has found its way into nearly every industry – from manufacturing to healthcare, travel, media and so many more. Even so, it’s easy to overlook the myriad ways that we interact with AI and ML in our daily lives, as so much of it happens behind the scenes

There’s still a ton of untapped potential when it comes to implementing machine learning, and we’re guaranteed to see even more applications in our everyday lives as time goes on.

If you’re curious how machine learning and AI can impact your business, we’ve compiled a free ebook with 50 use cases to help you get inspired.

Let’s block ads! (Why?)

RapidMiner

Analyze Data with Azure Machine Learning: No Coding Needed

July 3, 2020 CRM News and Info

Benign or Malignant? How AKA Used Azure Automated Machine Learning to Predict a Tumor—Without ANY Coding

Taking data and making it work for your company isn’t always an easy feat. As a Microsoft Dynamics 365 CRM technical consultant at AKA Enterprise Solutions, I’m constantly working with large databases of CRM data and it got me to thinking—what are the best solutions available for putting massive amounts of data to work? After a bit of research, I was introduced to Azure Automated Machine Learning from Microsoft. This solution was especially interesting to me because it was designed to be a no code (or low code) solution, compatible with datasets from an array of sources—including CRM, SQL, Blog storage and so on.

Why Choose Microsoft’s Azure Automated Machine Learning

Azure Automated Machine Learning takes a given dataset and swiftly trains machine learning models to discover relationships in this data. The key to Automated Machine Learning is its ability to see relationships that even experienced humans can’t. This technology enables businesses to gain a deeper understanding of their business as a whole and how clients interact with their company. Automated Machine Learning allows companies to take data and ultimately turn it into a better experience for their customers. This solution from Microsoft can also be deployed as a web service, so it can be used in other applications like Power BI, etc.

In the past machine learning was a new technology—both expensive and time consuming to implement. Microsoft was able to push past these barriers with the Azure Automated Machine Learning application, which is included with Azure. Although the CRM data analysis solution isn’t free, it is reasonably priced for what it delivers. Microsoft even provides users access to their project cost calculator to manage and plan costs before you jump in with both feet.

The time investment from start to finish varies, but I took a raw dataset from Breast Cancer Wisconsin (Diagnostic) and turned it into a fully trained machine learning model that can predict whether a tumor is benign or malignant in just one hour. It’s pretty amazing that this tool could take 569 digitized breast cancer tumor images and deliver their binary classification so quickly!

See just how I used Azure’s Automated Machine Learning Application to classify breast cancers tumors from a dataset in only 21 steps:

These types of results aren’t limited to healthcare—just imagine the possibilities for your industry. If your business wants more from your data, Microsoft’s Azure Automated Machine Learning may be the next step. You can learn more about this CRM data solution by reading Microsoft’s Azure Automated Machine Learning documentation. Or, contact AKA’s Azure experts and we will can help you get started or answer your questions.

ABOUT AKA ENTERPRISE SOLUTIONS
AKA specializes in making it easier to do business, simplifying processes and reducing risks. With agility, expertise, and original industry solutions, we embrace projects other technology firms avoid—regardless of their complexity. As a true strategic partner, we help organizations slay the dragons that are keeping them from innovating their way to greatness. Call us at 212-502-3900!

Article by: Nick Castro | 212-502-3900

Nick Castro is an Associate Technical Consultant at AKA Enterprise Solutions. With a strong background in Computer Science, Applied Mathematics and Statistics, and multiple Dynamics 365 certifications, he is able develop innovative and unique CRM solutions.

Let’s block ads! (Why?)

CRM Software Blog | Dynamics 365

Tag Archives: Machine

From DIY to open source

Massive progress with small data

The power of the collective

Key Performance Indicators for Overall Equipment Effectiveness

Key Benefits of Using Overall Equipment Effectiveness

Optimizing Overall Equipment Effectiveness with ML

Final Thoughts

Regression algorithms

1. Linear regression

2. ARIMA

Clustering algorithms

3. k-means clustering

4. Agglomerative & divisive clustering

Classification algorithms

5. k-nearest neighbors

6. Tree-based algorithms

7. Support vector machine

8. Neural networks

Other algorithm types

9. Fuzzy matching

10. Feature selection algorithms

Applications of Machine Learning in Everyday Life

1. Managing traffic

2. Making beer

3. Better Translation

4. Creating video games

5. Redefining music

6. Moderating social media

7. Deciding what to eat

8. Tutoring students

9. Helping to fight COVID-19

10. Developing new fragrances

Wrapping Up

Benign or Malignant? How AKA Used Azure Automated Machine Learning to Predict a Tumor—Without ANY Coding

Why Choose Microsoft’s Azure Automated Machine Learning

Recent Posts

Categories

Archives