vccafe

30 MACHINE INTELLIGENCE STARTUPS TO WATCH IN ISRAEL

Artificial Intelligence and Machine Learning will be eating the world. Don’t take my word for it – in the roundup of venture capital predictions for 2017, I found it to be the top recurring theme. Some examples:

  • AI will be the new mobile. Investors will ask management what their “AI strategy” is before investing and will be wary of companies that don’t have one (Fred Wilson, USV)
  • Security will shift from defensive to predictive AI-powered security (Norwest Venture Partners)
  • AI, M/L, And D/L will continue to be darlings of VC & M&A? (Chris Rust, Clear Ventures)
  • Artificial Intelligence is an important, foundational technology that gets more important every year and will be used to solve more and more problems going forward. Many large companies will be built (Michael Wolfe, Point Nine Capital)

Since AI and ML startups cut across verticals (analytics, fintech, health, adtech, security, etc), it’s easier to group them under the “machine intelligence” umbrella, coined by Shivon Zillis, a partner at Bloomberg Beta. In 2016 alone, 300+ “machine intelligence” (AI + ML) startups in Europe raised over €1.4 Billion in VC funding.

Venture Capital investments in AI, Europe 2016 (Source: Dealroom)

But what’s signal and what’s noise? Is “Deep Learning” the new “Big Data”? 

While the terms AI and ML get thrown around readily, the companies that truly apply Artificial Intelligence, Machine Learning, Computer Vision and Deep Learning have the potential to address problems that were unsolvable before.

DeepMind beating the world’s best at Go and AI beating pro-poker players consistently, represent much more than a computer getting good at playing video games:

The gaming world offers a perfect place to start machine intelligence work (e.g., constrained environments, explicit rewards, easy-to-compare results, looks impressive)—especially for reinforcement learning. And it is much easier to have a self-driving car agent go a trillion miles in a simulated environment than on actual roads. Now we’re seeing the techniques used to conquer the gaming world moving to the real world (Shivon Zillis).

These events mean that technology is advancing fast enough to make better decisions than humans in order to accomplish a given task. What tasks are next? that’s the essence of the list of startups below: the contenders who want to ultimately replace X with AI.

alphago deepmind google AI

Lee Se-dol (right), a legendary South Korean player of Go, poses with Google researcher Demis Hassabis before the Google DeepMind Challenge Match in Seoul. (source)

Noteworthy Machine Intelligence startups in Israel ??

There’s been a proliferation of Machine Intelligence companies in Israel. Below are 30 Artificial Intelligence (AI) and Machine Learning (ML) startups to watch in Israel (Part 1).

  1. Voyager Labs (2012) – cognitive computing for understanding human behaviour. Voyager’s cognitive-computing, deep-insights platform assesses billions of publicly available, unstructured data points to provide insights for its clients in finance, retail and consulting. The company emerged from stealth in November 2016 and announced a $100 million investment round from Sir Ronald Cohen, Lloyd Dorfman, OCAPAC Holding Company, and Horizons Ventures.
  2. Anodot (2014) – a real-time analytics and automated anomaly detection system that discovers outliers in vast amounts of data. Raised $8M in series B in September 2016. The company took an approach prevalent in cyber security and ported it to Business Intelligence.
  3. Beyond Verbal (2012) –  understands people’s moods, attitudes and emotional characteristics (also known as personality) from their raw vocal intonations in real-time, as they speak. One of the interesting applications for this technology is assessing intellect in job interviews.
  4. Fraugster (2014) – Technically based in Berlin, the company uses AI to predict malicious attacks before they happen. This is done by enriching transaction data points such as name, email address, and billing and shipping address with around 2,000 extra data points, such as an IP latency check to measure the real distance from the user, IP connection type, distance between key strokes, and email name match. Chen Zamir is the company’s CTO and former Intelligence officer in the IDF as well as Paypal risk manager.
  5. Innoviz Technologies (2016)- High Definition Solid State LiDAR (HD-SSL), enables smart and advanced 3D remote sensing for fully autonomous vehicles, while significantly reducing both cost and size. All key technologies for autonomous driving. Innoviz recently teamed up with Magna, a tier-1 supplier to the automotive industry.
  6. Kang Health (2016) – HQ’d in New York and started by Allon Bloch, the former CEO of Wix.com, Kang Health wants to crowdsource health data to provide better information about health conditions, symptoms and treatments. the company raised a strong $3.3M in seed round in November.
  7. Orcam (2010)– Orcam can potentially help the 280 million visually impaired people to automatically read any text they are looking at via a discreet device that attaches to the user’s eyeglasses. Orcam was created by the founders of Mobileye (NYSE: MBLY and one of Israel’s largest IPOs), Prof. Amnon Shashua and Ziv Aviram (great interview with them here). The company unveiled its latest model in January 2017 at CES, offering the blind the ability to read and recognize faces.
  8. Twiggle (2014) – Twiggle develops next generation e-commerce search leveraging advanced techniques in data science, artificial intelligence, machine learning and natural language processing (NLP). As part of their recent investment round by Alibaba, Twiggle’s announced Udi Manbar, former head of search at Google and Chief Scientist at Yahoo joined the company’s board of directors.
  9. Windward (2010) – maritime data and analytics company. Windward processes more than 100 million data points every day to deliver unprecedented insights into the vital cargo-shipping industry as well as security (smuggling, sovereignty etc). The platform does this through combining ships’ location data with data about the ships (their capacity to carry weight vs. their actual weight via satellite images) to map optimized shipping paths and behaviours while at sea. Co-founders Ami Daniel and Matan Peled both served as naval officers. This interview with Ami sheds more light on the company.
  10. YouAppi (2011)- YouAppi  creates adtech to streamline mobile user acquisition. The company’s tech predicts the right app and location to present an ad based on user and cohort behaviour. It uses machine learning and predictive algorithms to analyze over 250 terabytes of data daily.
  11. Chorus.ai (2015) –  Chorus uses AI to analyze sales calls by joining the conference calls, transcribing the data and extracting important action items. The company just raised its $16M series A led by Redpoint Ventures, just four months after announcing their seed funding.
  12. Cimagine (2012) – launched the world’s largest implementation of AR in retail to date with Shop Direct. Rumoured to have been acquired by Snap in December 2016 for $30-40M and will likely serve as Snap’s R&D center in Israel.
  13. Zebra Medical Vision – The company combines its vast imaging database with deep learning techniques to build algorithms that will automatically detect and diagnose medical conditions – helping radiologists to detect overlooked indications and give fast, accurate imaging diagnosis. Zebra Medical Vision was recently selected as one of the most Innovative AI companies for 2017 by Fastcompany.
  14. Logz.io (2014) – Logz.io is an AI-powered log analysis platform that offers the open source ELK Stack as an enterprise-grade cloud service with machine learning technology. Provides real-time access to data insights based on the collaborative knowledge of system administrators, DevOps engineers, and developers throughout the world. The company completed its $16M Series B in November 2016.
  15. Revuze (2011) – The company graduated from the Nielsen Innovate incubator in Caesarea, the company analyzes user sentiment on products, product-attributes and brands by analyzing reviews as well as sentiment extraction from survey responses, call-center text and social media. Source text is analyzed against category taxonomies generated via semi-supervised machine learning.  They are part of a bigger trend to automatically evaluate user feedback.
  16. Fifth Dimension (2014) – Fifth dimension provides big data analysis for the Intelligence and security space. It can process petabytes of data in real time to identify faces in crowds, voices in audio recordings and predict threats, (and opportunities) before they become a reality. The company’s chairman was the former General Chief of Staff of the IDF 2011-2015.
  17. Deep Instinct (2014)- Deep Instinct safeguards the enterprise’s end-points or mobile devices against any threat on any infrastructure, whether or not it is connected to the network or internet. Deep Instinct was named “Cool Vendor in 2016” by the Gartner Research Group.
  18. Nexar (2015)-  A community-based AI dashboard cam app that helps drivers to protect themselves on the road and provides documentation, recorded video, and situational reconstruction in case of an accident. The Nexar solution employs machine vision and sensor fusion algorithms, leveraging the iPhone’s sensors to analyze and understand the car’s surroundings. Nexar recently hired deep learning heavyweight Professor Trevor Darrell from UC Berkeley as Nexar’s Chief Scientist.
  19. SparkBeyond (2013) – an automated general purpose research engine designed to leverage and intelligently augment masses of data that exist on the web, and discover complex patterns within them. The SparkBeyond Discovery Platform is being used by Fortune 500 companies in the Finance, Manufacturing, Life-Sciences, Energy, e-Commerce, Internet and Healthcare industries.
  20. Fdna (2010) – FDNA stands for Facial Dysmorphology Novel Analysis, a technology that transforms facial photos into deep and accurate phenotypic information in real time. The company developed the Face2Gene suite of phenotyping apps that facilitate comprehensive and precise genetic evaluations through computer vision.
  21. Loom systems (2015)– Loom Systems Ops sends proactive notifications about meaningful issues in an IT environment, empowering its end-users with visibility into their IT blindspots. The company applies AI to sculpt big data into reports delivered in plain English.
  22. AIdoc (2015)- Aldoc applies deep learning to the radiology space by guiding the radiologist to the most relevant places in the scan and consolidating multiple sources of data to one screen. The company hopes to take the number of diagnostic errors to zero. The company recently completed its $3.5M series A in November of 2016.
  23. Atidot (2016) – an InsureTech startup that offers a SaaS platform to insurers. Atidot offers a predictive analytics platform for actuarial science and risk management.
  24. Cognata (2017) – still in stealth, Cognata is developing Artificial Intelligence simulation engines for automated vehicles. The company has no public URL yet.
  25. Neurala (2006) – Neurala’s Brains for Bots SDK helps bring artificial intelligence to drones, robots, cars, and consumer electronics by helping these devices inspect their environment, make decisions and navigate obstacles. The company already works with a broad range of clients including the US Air Force, Motorola and Parrot; to back its vision it recently raised $14M series A in January 2017.
  26. Javelin Networks (2014) – the company’s main value proposition is to protect the Active Directory (used by 9 out of 10 companies) from cyber attacks by combining AI, obfuscation and advanced forensics methodologies right at the point of breach. Javelin has just announced its $5M series A in February 2017.
  27. Dynamic Yield (2011)- Dynamic Yield’s advanced machine learning engine builds actionable customer segments in real time, enabling marketers to increase revenue via personalization, recommendations, automatic optimization & 1:1 messaging. Media and e-commerce sites are prime customers of this. The company has just announced its $22M series C in December 2016.
  28. Dragonera (2016) – AI based software development service. The platform can supposedly automate up to 70% of early development of new products by leveraging micro-services, and pre-existing pieces of code. The timeframe for a fully functional product varies between 14 and 45 days. Dragonera raised $3M seed round in December led by Singulariteam.
  29. MedyMatch Technology (2013)- automated medical diagnostic support system that utilizes machine learning and expert feedback to deliver diagnostic recommendations for medical imaging. Raised $2M in seed funding in March 2016.
  30. Augury (2012) – predictive maintenance for industrial IOT. Augury automatically diagnoses machines based on the sounds they make. The product connects vibration and ultrasonic sensors to smartphones and pairs them with machine-learning algorithms to reduce environmental impact, energy usage, and operational costs. See Augury’s tech in action on The Verge.

Credit is due to Ha Duong who compiled a list of 600+ AI startups in Europe at Tech.eu: 600+ European AI tech startups to watch, which served as a starting base for this list. On the next post, you’ll see 20+ more Israeli startups in Machine Intelligence and the clusters that are being formed. Know a company that needs to be on this list? drop me a line at eze at vccafe dot com.

source: http://www.vccafe.com/2017/02/15/30-machine-intelligence-startups-to-watch-in-israel/

Read More

vb logo

AI startups that analyze customer reviews

Already, as of 2010, a quarter of Americans (24 percent) had posted product reviews or comments online, and 78 percent of internet users had gone online for product research. But those are ancient stats. Numbers are higher now. More recently, BrightLocal found in 2016 that 91 percent of consumers regularly or occasionally read online reviews, with 47 percent taking sentiment of local-business reviews — the tonality of a review’s text — into account in purchasing decisions. Breaking out the figures, 74 percent of consumers say that positive reviews make them trust a local business more, and 60 percent say that negative reviews make them not want to use a business, according to BrightLocal.

So reviews are important, and the feelings expressed are key. To understand review content, including sentiment, at web and social scale and velocity, you need automated natural language processing (NLP) and other forms of AI.

Commercial review-management platforms — from Bazaarvoice, PowerReviews, Yotpo, and others — help brands and online commerce sites collect reviews and redeploy them to boost sales. That’s an important function: Bazaarvoice reports, “Our clients see 65% average lift in revenue per visit and 52% lift in conversion on product pages with ratings and reviews.” Not all platforms bake NLP into their products and services, however. It’s the companies that do that interest me, the ones that look at what’s actually said in reviews, beyond the star ratings. Let’s look at four, and then at do-it-yourself approaches to customer-review analysis.

Customer reviews contain several forms of salient information. First there’s the star rating, but ratings, even when broken out into categories — on Airbnb, for example, categories include accuracy, communication, cleanliness, location, check-in, and value — have zero explanatory power. So we have review text: free-form, voice-of-the-customer reactions. This text tells a story, and stories sell, so we need to know the aspects of a product or service discussed, the wording used to describe them, and the sentiment expressed.

Review text also reveals a lot about the reviewer, as Stanford University Prof. Dan Jurafsky explains in an exploration of review language, Natural Language Processing on Everyday Language. (Jurafsky’s data science study on how restaurants and reviewers talk about food — including the connection between menu wording and item price — is really illuminating. Also, NLP and AI can help in review moderation by identifying abusive language and detecting fraudulent reviews, but those are topics for another article.) Finally, reviewer identity is key: Demographic characteristics such as age, gender, and geographic location, as well as reviewer reputation or ratings and the reviewer’s social profile, come into play, as does review recency.

We’re describing a complex data scenario. A Bazaarvoice blog post will take you through some of the data science challenges, but it doesn’t cover solutions. The startups I will profile deploy analytics — NLP, machine learning, and other forms of AI — to respond to the challenges.

Revuze

Above: The Revuze analysis dashboard

1. Revuze focuses on products and product attributes in addition to brand health, with a couple of differentiators. One is special attention to discovering different ways people talk about a given topic, and a second is the ability to identify sentiment in phrases that lack obvious clues like using words like “good,” “happy,” and “terrible.”

Revuze analyses aren’t limited to reviews; the company’s tech applies applies NLP for topic, keyword, and sentiment extraction from survey responses, call-center text, and social media too. Source text is analyzed against category taxonomies generated via semi-supervised machine learning. The company graduated from the Nielsen Innovate incubator in Caesarea, Israel in the fall of 2015 and has turned its attention to product experience management (PEM), enabling clients to “measure customer perception of the holistic product and service experience.”

Aspectiva

Above: Aspectiva’s website-embedded Crowd Opinions Overview

2. Aspectiva provides aspect-based review aggregation and product search, focused on reviewers’ perceptions of product attributes and capabilities. Aspect extraction, implemented via unsupervised machine learning, is coupled with behind-the-scenes analytics to reveal “the true uses of any product and generates recommendations based on the full user experience.”

A results graphic — product aspects and sentiment ratings, as shown in the image — can be embedded in an online commerce site. The goal, per the Brandwatch report cited above, is to boost conversion rates and sales revenue. Aspectiva also provides an API, allowing customers to build their own front-ends to Aspectiva analytics, and a search function that is cognizant of product attributes.

Aspectiva deploys an NLP–machine learning combination that “scans texts written by consumers and learns what people are saying when they are happy or unhappy with products they write about, beyond the obvious ‘sentiment words’ themselves … determining sentiment also with factual sentences.” This capability isn’t unique to Aspectiva — Revuze claims something similar — but it makes for more robust sentiment detection than is found in many competing products.

Bildschirmfoto 2017-01-10 um 13.58.11

Above: SmartMunk’s Customer Satisfaction Dashboard

3. SmartMunk story.ly analysis centers on satisfaction drivers rather than on product and service attributes. The principle is that customer satisfaction drives business outcomes, so it’s important to focus on elements that make customers happy or that disappoint.

The company’s hybrid methodology quantifies qualitative insights discovered in consumer-generated content, targeting brand product development and marketing functions. “story.ly directly gathers your reviews including filter variables from seller platforms. Within seconds you see the story in your smart online report.”

I especially like the skyline ontology graphic seen in the lower-right corner of the Customer Satisfaction Dashboard. SmartMunk founder Andera Gadeib, who also runs market-research agency Dialego, points out the functional and emotional attributes captured in the ontology, a categorical representation of satisfaction drivers. As for the theme cluster: It is generated via TF-IDF term ranking, based on the relative frequency of term occurrence within the set of inputs. According to Gadeib, although “clients like to look at coded data” — at consumer-generated content classified according to product attributes (rather than satisfaction drivers) — the company “has had good success going with a less category-driven view.”

SentiGeek Architecture

Above: SentiGeek’s review-analysis technical architecture

4. SentiGeek is a pre-launch customer feedback/review-sentiment company. (I informally advise founder Mara Tsoumari, who presents features and possibilities in an online video.) Candidate markets — indicating the breadth of review-sentiment interest — include online retail, market research, marketing, financial institutions, and public administrations. The product is designed to support reporting, analysis, and monitoring options. It extracts opinion words and phrases and fine-grained sentiment and provides analyses by review and by opinion holder, with the ability to generate customer profiles.

It’s the reviewer-focused analyses that differentiate SentiGeek, but rather than provide an interface screenshot for illustration, I thought it would be more interesting to look at SentiGeek’s technical architecture. Note the use of spaCy, an innovative, open source NLP package that features parsing and named-entity extraction capabilities. Note also the use of an RDF-structured knowledgebase — RDF is the Resource Description Framework, which originated as a Semantic Web standard — designed for domain-specific aspect–sentiment evaluations.

Do-it-yourself choices

Do-it-yourself is an option if you have strong data-wrangling and analytical skills. DIY option #1 is to analyze review text within a data analysis workbench: for example, Aylien describes Building a Text Analysis process for customer reviews in RapidMiner, and MeaningCloud covers Text Classification in Excel: build your own model. If you have data science skills, Python is a great choice, perhaps using gensim, NLTK, Stanford CoreNLP (via a Python wrapper), or TensorFlow. You can use spaCy NLP — SentiGeek’s choice — for Python native information extraction; spaCy plays nice with TensorFlow, Keras, Scikit-Learn, Gensim, and the rest of Python AI ecosystem.

You can also build it yourself, applying a commercial NLP service. Hearing John Kelley describe his work at TripAdvisor (using tech from Lexalytics) is what first opened my eyes to the possibilities that review analytics provide. Video of Kelley’s presentation, What Travelers Say… Using Sentiment to Improve User Engagement, is dated but still quite interesting.

A look ahead

Expect consumers to continue sharing product and service perceptions, focusing on both features and experiences. “Word of mouth” matters: Influence on purchase decisions, of reviews and social postings, will likely only grow. We will see a continued strong market for analytics — NLP, machine learning, and data science — as a best response to the volume–impact combination, text analytics (and also image and video analytics) that can detect entities, emotions, and context in diverse media. If you’re a brand, you need to build or adopt an analytics solution, whether from an established vendor or a startup, or you risk falling behind.

 

Surce: VB

Read More

breakthrough analysis

HOW FOUR AI STARTUPS HELP BRANDS EXPLOIT CUSTOMER REVIEWS

Already, as of 2010, a quarter of Americans (24%) had posted product reviews or comments online, and 78% of Internet users had gone online for product research. But those are ancient stats. Numbers are higher now. More recently, BrightLocal found in 2016 that 91% of consumers regularly or occasionally read online reviews, with 47% taking sentiment of local-business reviews — the tonality of a review’s text — into account in purchasing decisions. Breaking out the figures, 74% of consumers say that positive reviews make them trust a local business more and 60% say that negative reviews make them not want to use a business, according to BrightLocal.

So reviews are important, and the feelings expressed are key. To understand review content including sentiment, at Web and social scale and velocity, you need automated natural language processing (NLP) and other forms of AI.

Commercial review-management platforms — from Bazaarvoice, PowerReviewsYotpo, and others — help brands and online commerce sites collect reviews and redeploy them to boost sales. That’s an important function: Bazaarvoice reports, “Our clients see 65% average lift in revenue per visit and 52% lift in conversion on product pages with ratings and reviews.” Not all platforms bake NLP into their products and services, however. It’s the companies that do that interest me, the ones that look at what’s actually said in reviews, beyond the star ratings. Let’s look at four, and then at do-it-yourself approaches to customer-review analysis.

Four AI startups that help brands exploit customer reviews

Customer reviews contain several forms of salient information. First there’s the star rating, but ratings, even when broken out into categories — on Airbnb, for example, accuracy, communication, cleanliness, location, check-in, and value — have zero explanatory power. So we have review text: free-form, voice of the customer reactions. This text tells a story, and stories sell, so we need to know the aspects of a product or service discussed, the wording used to describe them, and the sentiment expressed.

Review text also reveals a lot about the reviewer, as Stanford University Prof. Dan Jurafsky explains in an exploration of review language, Natural Language Processing on Everyday Language. (Jurafsky’s data science study how restaurants and reviewers talk about food — including the connection between menu wording and item price — is really illuminating. Also, NLP and AI can help in review moderation by identifying abusive language and can detect fraudulent reviews, but those are topics for another article.) Finally, reviewer identity is key, demographic characteristics such as age, gender, and geographic location and also reviewer reputation or ratings and the reviewer’s social profile, and other factors come into play such as review recency.

We’re describing a complex data scenario. A Bazaarvoice blog post will take you through some of the data science challenges but it doesn’t cover solutions. The startups I will profile deploy analytics — NLP, machine learning, and other forms of AI — to respond to the challenges.

revuze-dashboard-sh3
The Revuze analysis dashboard.

Revuze focuses on products and product attributes in addition to brand health, with a couple of differentiators. One is special attention to discovering different ways people talk about a given topic, and a second is ability to identify sentiment in phrases that lack obvious clues, that don’t use words like “good,” “happy,” and “terrible.”

Revuze analyses aren’t limited to reviews; the company’s tech applies applies NLP for topic, keyword, and sentiment extraction from, additionally, survey responses, call-center text, and social media. Source text is analyzed against category taxonomies generated via semi-supervised machine learning. The company graduated from the Nielsen Innovate incubator in Caesarea, Israel in the fall of 2015 and has turned its attention to Product Experience Management (PEM), enabling clients to “measure customer perception of the holistic product and service experience.”

Technically similar to Revuze but with a very different orientation —

aspectiva
Aspectiva’s Web-site embedded Crowd Opinions Overview.

Aspectiva provides aspect-based review aggregation and product search, focused on reviewers’ perceptions of product attributes and capabilities. Aspect extraction, implemented via unsupervised machine learning, is coupled with behind the scenes analytics to reveal “the true uses of any product and generates recommendations based on the full user experience.”

A results graphic — product aspects and sentiment ratings as shown in the image at right — may be embedded in an online commerce site. The goal, per the Brandwatch report cited above, is to boost conversion rates and sales revenue. Aspectiva also provides an API, allowing customers to build their own front-ends to Aspectiva analytics, and a search function that is cognizant of product attributes.

Aspectiva deploys an NLP–machine learning combination that “scans texts written by consumers and learns what people are saying when they are happy or unhappy with products they write about, beyond the obvious ‘sentiment words’ themselves… determining sentiment also with factual sentences.” This capability isn’t unique to Aspectiva — Revuze claims something similar — but it makes for more robust sentiment detection than is found in competing products.

SmartMunk's Customer Satisfaction Dashboard
SmartMunk’s Customer Satisfaction Dashboard

SmartMunk story.ly analysis centers on satisfaction drivers rather than on product and service attributes. The principle is that customer satisfaction drives business outcomes so it’s important to focus on elements that make customers happy or that disappoint.

The company’s offers a hybrid methodology that quantifies qualitative insights discovered in consumer generated content, targeting brand product development and marketing functions. “story.ly directly gathers your reviews including filter variables from seller platforms. Within seconds you see the story in your smart online report.”

I especially like the skyline ontology graphic seen in the lower-right corner of the Customer Satisfaction Dashboard at right. SmartMunk founder Andera Gadeib, who also runs market-research agency Dialego, points out the functional and emotional attributes captured in the ontology, a categorical representation of satisfaction drivers. As for the theme cluster: It is generated via TF-IDF term ranking, based on the relative frequency of term occurrence within the set of inputs. According to Andera Gadeib, although “clients like to look at coded data” — at consumer-generated content classified according to product attributes (rather than satisfaction drivers) — the company “has had good success going with a less category-driven view.”

SentiGeek's review-analysis technical architecture
SentiGeek’s review-analysis technical architecture

SentiGeek is a pre-launch customer feedback/review-sentiment company. (I informally advise founder Mara Tsoumari, who presents features and possibilities in an online video.) Candidate markets — indicating the breadth of review-sentiment interest — include online retail, market research, marketing, financial institutions, and public administrations. The product is designed to support reporting, analysis, and monitoring options. It extracts opinion words and phrases and fine-grained sentiment and provides analyses by review and by opinion holder with the ability to generate customer profiles.

It’s the reviewer-focused analyses that differentiates SentiGeek, but rather than provide an interface screenshot for illustration, I thought it would be more interesting to look at SentiGeek’s technical architecture, shown in the image at right. Note the use of spaCy, an innovative, open-source NLP package that features parsing and named-entity extraction capabilities. Note also the use of an RDF-structured knowledgebase — RDF is the Resource Description Framework, which originated as a Semantic Web standard — designed for domain-specific aspect–sentiment evaluations.

Do-it-yourself choices

Do-it-yourself is an option if you have strong data-wrangling and analytical skills. DIY option #1 is to analyze review text within data analysis workbench. Examples: Aylien describes Building a Text Analysis process for customer reviews in RapidMiner and MeaningCloud covers Text Classification in Excel: build your own model. If you have data science skills, Python is a great choice, perhaps using gensim, NLTK, Stanford CoreNLP (via a Python wrapper), or TensorFlow. (Links are resources and examples.) Use spaCy NLP — SentiGeek’s choice — for Python native information extraction; spaCy plays nice with TensorFlow, Keras, Scikit-Learn, Gensim, and the rest of Python AI ecosystem.”

Build-it-yourself, applying a commercial NLP service, is another option. Hearing John Kelley describe his work at TripAdvisor (using tech from Lexalytics) is what first opened my eyes to review-analytics possibilities. Video of Kelley’s presentation, What Travelers Say… Using Sentiment to Improve User Engagement, is dated but still quite interesting.

A look ahead

Expect consumers to continue sharing product and service perceptions, focusing on both features and experiences. “Word of mouth” matters: influence on purchase decisions, of reviews and social postings, will likely only grow. We will see a continued strong market for analytics — NLP, machine learning, and data science — as a best response to the volume–impact combination, text analytics and also image and video analytics that can detecting entities, emotions, and context in diverse media. If you’re a brand, build or adopt an analytics solution, whether from an established vendor or a startup, or fall behind.


For more on review Voice of the Customer analyses and related topics, check out the 2017 Sentiment Analysis Symposium, taking place June 27-28 in New York, tagline Emotion–Influence–Activation. We have a Call for Speakers open through January 31.

source: Breakthrough Analysis

Read More

CB INSIGHTS

11 Early-Stage Israeli B2B Software Companies To Watch

Israel is one of the key tech markets globally, consistently ranking among the top 10 countries in our semi-annual tech exits report. While the market is known for innovation in cybersecurity and business intelligence, Israeli entrepreneurs are also launching companies in other pockets of tech such as auto tech and industrial IoT.

We used CB Insights Mosaic scoring tool, which uses public data and predictive algorithms to measure the overall health and growth potential of private companies, to identify 11 early-stage companies in the seed and Series A stage with traction.

We specifically looked at business-to-business companies that have all raised funding since January 2016, but have not yet raised a Series B round.

Among the startups on our list is Indegy, a cybersecurity company that has raised roughly $18M to date, making it the most well-funded company on our list. In second place is auto tech company Oryx Vision, with $17M in funding. Following behind, two companies on this list—Otonomo, Alooma, —have raised $15M in total funding.

Cybersecurity is the most popular category, with three companies working in the field. Other categories include auto tech, ag tech, business analytics, and even one company working on restaurant tech.

Check out the list below.

CHART

Source:  CB Insights

Read More

pymnts blog

Meet Revuze, The Startup Bringing Artificial Intelligence To Brand Management

 

Revuze is hoping to bring artificial intelligence to brand management.  http://www.pymnts.com/news/merchant-innovation/2016/revuze-startup-artificial-intelligence-brand-management/

Let’s face it, data analysis can be a costly and time-consuming affair and can sometimes take weeks or months of scouring through troves and troves of data from a number of channels, like sales, surveys, customer reviews and social media, to even interpret how consumers might be responding to a certain brand or product.

Then, of course, there’s always the chance that human error or bias could lead to the wrong interpretation or understanding of the data as well.

But what if there was a faster, cheaper, quicker and more efficient way to crunch all that data to find out exactly how consumers think or feel about a brand?

That’s exactly what Revuze, an Israeli-based startup, is aiming to do by using an AI powered by neural networks and machine learning to cut the human element and timely and costly data crunching out of brand and product management by offering up almost-immediate insight into how a brand’s products and services are being experienced by the consumer.

“Industries … have previously relied on manually intensive solutions, such as analytics, social listening and monitoring,” according to Revuze. “These current solutions, requiring months to execute, demand teams of product experts, data scientists and analysts to construct and maintain rules, dictionaries and taxonomies before interpreting the findings. This process is slow, expensive, inefficient and easily skewed by misunderstanding or misinterpretation. Revuze removes these limitations, directly addressing the biases and other shortcomings of those managing such extreme volumes of data.”

Revuze, headquartered in Netanya, Israel, announced this week that it raised $4 million in seed funding from strategic investors Nielsen, The NPD Group and TIC Group, which will allow it to expand by opening U.S. offices in San Francisco and New York City. The funding will also allow Revuze to introduce its AI-based brand intelligence technology to its investors’ customers, according to Cofounder and CEO Ido Ramati.

“Global brands have told us that this is a game-changer for managing their brand health,” Ramati said. “They are now able to make the kind of key business decisions and utilize data from such varied sources in ways that before were not possible. Through our introduction of AI into this field, brands can now gain an immediate, granular understanding of what customers are saying about their brand, product or competitors, without having to hire teams of experts.”

The way Revuze’s AI-based tech works is pretty simple. First, Revuze scans a massive amount of customer feedback across multiple internal and external sources — think online product reviews, social media, sales reports, call centers, customer emails, surveys and more — in multiple languages to find out exactly what consumers think about their experience with a brand’s products and services.

Ramati said the AI-based tech is like having a 24/7 “analyst in a box” because it allows brands to really delve into their own customer data and figure out what questions they should be asking about how/why consumers feel a certain way about their products.

Plus, the software is so easy to use that pretty much any employee at any level of the company can utilize it to help the brand’s growth and development.

“With a simple one-time Q&A session — and without the need to pre-define a single keyword, rule, topic or sentiment value — a junior employee, in days, can arrange to have Revuze deliver the most nuanced information available,” according to Revuze’s statement announcing its funding. “The data is delivered, via a single screen, about consumers and their needs, across thousands of internal and external data sources on any given product family.”

If Revuze can take the human element out of data crunching and deliver quicker more insightful information about how consumers really feel about a brand’s products and services, it should be a very interesting AI developer to watch in the future.

“Revuze is transforming the way that companies track their corporate and product brands through the introduction of AI,” John Burbank, Nielsen’s president of strategic initiatives, said in a statement. “Now, they can easily gain detailed market and consumer intelligence, which enables them to be far more responsive to consumer perception and feedback.”

By PYMNTS

Posted on September 6, 2016

Read More

wall

Revuze Raises $4M Seed for Brand Performance Data-Mining

By Patience Haggin

Aug. 31, 2016 7:30 a.m. ET

The venture capital arm of Nielsen Holdings PLC, a company that provides businesses with data about consumers, has led a $4 million seed round in a startup that uses artificial intelligence to gather data on brands’ performance on product and customer satisfaction.

Nielsen Innovate invested in Revuze, a company mines unstructured data sources including social-media posts and reviews on e-commerce websites to collect data on a brand’s customer satisfaction. The company’s funding round also included NPD and TIC Group.

“It’s about extracting from this gigantic sea of data that’s publishing every day to help brands better understand their marketing and product,” said John Burbank, president of strategic initiatives at Nielsen. “We’re bringing Revuze out to our clients. We almost treat them as if they’re part of the Nielsen family. What happens down the road is yet to be determined.”

Revuze will use the new funding to expand its sales and to build new features for the product. The company, founded in Netanya, Israel, will open offices in New York and San Francisco. It will look to raise another round around the end of this year or the beginning of next year, said co-founder and Chief Executive Ido Ramati.

Write to Patience Haggin at patience.haggin@wsj.com

Read More

vator2

A quick chat with Ido Ramati, co-founder and CEO of Revuze

Financial trends and news by Steven Loeb
August 31, 2016

In the world of business, companies have huge volumes of unstructured feedback pouring in via call centers, emails, surveys, charts, online comments, blogs and social media. That makes it extremely difficult to gleam any actionable insights.

That’s the problem that Revuze is looking to solve. It uses artificial intelligence-powered technology that enables brands to better understand product, brand management and customer satisfaction issues. all while streamlining the Product Experience Management and automatically scoring and ranking a brand’s performance, both relative to its competitors and to the market as a whole.

On Wednesday, the company announced a $4 million seed-funding round led by Nielsen, The NPD Group, and TIC Group.

Founded in 2013, the Netanya, Israel-based company, which also operates in Australia, will use the new funding to expanding its U.S. operations by opening offices in San Francisco and New York City, where it plans to 10 additional employees by the end of next year. That would be on top of the 20 employees the company has right now.

I spoke to Ido Ramati, co-founder and CEO of Revuze, about the company and how it uses artificial intelligence to solve an age old problem.

VatorNews: What is the problem that Revuze is trying to solve? How does it solve it?

Ido Ramati: Revuze takes any amount of unstructured data. It could be from any industry, from fast food to automobiles to hotels. The first step, and this is where the artificial intelligence, neural networks and machine learning engines come in, is Revuze does not rely on the traditional method of hiring teams of data scientists, product and BI experts. Typically, they would manually predefine, construct and maintain rules, dictionaries and taxonomies before interpreting the findings, which results in a process that is very slow, expensive, inefficient and can lead to misinterpretation.

Revuze removes these limitations, directly addressing the biases and other shortcomings of those managing such extreme volumes of data, by automating the entire process and letting the system teach itself what and how to analyze the data. For example, instead of having an analyst predefine keywords and dictionaries that they think are important to measure, the Revuze platform generates and independently maintains a list of metrics to measure, based on the discussion topics it identified in the data.

The entire data from any of the sources is delivered, via an intuitive, easy-to-use dashboard, which presents all data sources in a single screen, rather than on seven or eight different screen or user interfaces with different solutions for different sources.  Moreover, Revuze allows its clients to benchmark any feature, functionality or experience of their products and brands to the industry benchmark, or to other competitive products or brands.

VN: Who is the typical customer for Revuze? Walk me through a typical use case

IR: Revuze caters to consumer brands of manufacturing companies, such as CPGs, consumer electronics and toys, and to service-oriented companies, such as fast-food chains, hospitality and banks. Revuze users are from fairly varied sections of an organization, such as marketing, R&D/product development, operations, BI and customer service.

Marketing teams, for example, are using our solution to replace their existing methods for tracking their brand health, to measure product launch success, to conduct a competitive analysis including strengths, weaknesses, opportunities, and threats (SWOT) analysis, and to understand the consumer buying criteria. R&D teams are using the solution to understand which product features are working properly and which features are getting too many complaints, and to understand which product metrics are important to their consumers.

VN: How many customers do you currently have? How are you growing?

IR: We have six customers. Strategic partners such as Nielsen and The NPD Group are introducing Revuze to their clients, who often end up working closely with us.

VN: Why is what Revuze does valuable to its customers? Have you calculated any ROI?

IR: Revuze provides its customer an intuitive, easy-to-use dashboard that replaces the need of using six to eight different solutions operated by teams of data scientists, product experts and analysts. We enable our clients to get immediate actionable insights from any source on three difference levels: Industry, or measuring the industry benchmark on every aspect; brand, which means comparing your brand performance to competing brands; and product, which involves analyzing down to a specific product SKU.

VN: Where would you like to see the company in the next five years?

IR: Our strategy is to continue our expansion both inside and outside the U.S. market, including adoption in different geographies by as many industries as possible. In addition, in the next five years we will continue to develop our groundbreaking technology which revolutionizes the text analytics space in a way that caters to other domains beyond the market research domain.

VN: Is there anything else I should know about the company?

IR: We are happy to see that the Fortune 500 companies we are working with, which have already lost faith in using text analytics because of their complexity and the vast resources they require, are regaining their trust via Revuze. Now, they have a better way to create better products and services and to interact with their clients.

We believe that in the same quick way the market research domain embraced our technology, other domains with different use cases will do the same.

 

Read More

marketingland3

AI-powered Revuze enters US market to help brands figure out what’s bothering customers

Customers have an endless supply of opinions about your products and services, and marketers are eager to boil them down into useful insights.

This need has led to a variety of sophisticated tools, including text/sentiment analysis fromLexalytics, “voice of the customer” analysis like Medallia, predictive analysis in Quantifind, contextual analysis from BehaviorMatrix or customer intelligence provider Clarabridge.

Now, another platform is entering the US market, with the promise that its neural networks and machine learning is smarter and more useful than the others.

Revuze — a Netanya, Israel-based firm with offices in New York City and, soon, San Francisco — describes its tool as a “brand health indicator.” It pulls in qualitative data from thousands of brand and third-party sources, including social media, customer surveys, website behavior, e-commerce activity and comments and emails to customer service.

The platform then analyzes all that customer data to determine, say, that the volume control on your latest smartphone model is hard to use.

While a variety of other tools can also yield such insights, Revuze Co-Founder and CEO Ido Ramati told me that a key differentiator for Revuze is that the brand doesn’t have to hand-hold the process.

In other words, instead of setting the tool to look for comments about volume control on this particular model, the brand just lets Revuze find out what concerns are rising to the top. There’s a one-time Q&A session, but no definitions of keywords, rules or topics, and no setting of sentiment value.

The system learns what is most important as the brand selects the results in several rounds that deliver lists of discussion topics and related data. Ramati says a key difference from, say, sentiment analysis is that the Revuze platform doesn’t require that humans sift the results to determine which feedback is good or bad, as the system learns what matters.

A cell phone manufacturer, for instance, might otherwise need a team of analysts to sift through analytical results in order to find that customers are unhappy with a particular model’s battery life, Ramati said, while Revuze’s engine “knows how to figure that out, [to pick] what customers care about.”

Revuze-Dashboard-SH3

The results can then be grouped by industry to provide a benchmark comparing feedback from competitors’ customers, or they can be grouped by the brand and the specific product model. Once the insights rise to the top, the brand can then choose to use them to improve a product design or a service — or not.

Ramati said that an unnamed global fast food client was receiving about a million customer surveys every month but couldn’t properly analyze them because of the complexity and volume. Revuze’s platform was called in, and he said it surfaced the top customer complaints about the speed of service and portion size in specific restaurants.

Founded in 2013, Revuze beta-tested the platform last year, and issued a release version by year’s end. The company said it has been employed by a variety of Fortune 500 brands in consumer product goods (CPG), food service, leisure and manufacturing but declined to name them.

Along with the US launch, the company is announcing a $4-million seed round from investors that include research firm Nielsen’s tech incubator Innovate.

Read More

yahoof

Revuze Introduces Artificial Intelligence to Transform Brand Intelligence, Raises $4 Million Seed Round, Opens U.S. Operations

Revuze, the first company to bring Artificial Intelligence (AI) to both brand and product management, is announcing a $4 million seed-funding round led by strategic investors Nielsen, The NPD Group, and TIC Group. Revuze is also entering into business development partnerships to introduce its transformative AI-led technology to its investors’ customers. On the back of this investment, the company is expanding its U.S. operations by opening offices in San Francisco and New York City.

Revuze uses AI, powered by neural networks and machine learning, to empower the brand and product management industries that previously have relied on manually intensive solutions, such as text analytics, social listening and monitoring. These current solutions, requiring months to execute, demand teams of product experts, data scientists and analysts to construct and maintain rules, dictionaries and taxonomies before interpreting the findings. This process is slow, expensive, inefficient and easily skewed by misunderstanding or misinterpretation. Revuze removes these limitations, directly addressing the biases and other shortcomings of those managing such extreme volumes of data.

With a simple one-time Q&A session – and without the need to pre-define a single keyword, rule, topic or sentiment value – a junior employee in days can arrange to have Revuze deliver the most nuanced information available. The data is delivered, via a single screen, about consumers and their needs, across thousands of internal and external data sources, on any given product family. Revuze has created and is widely introducing its technology available for Product Experience Management, enabling brands to quickly understand product and customer satisfaction issues, and to automatically score and rank the brand’s performance, both relative to its competitors and to the market as a whole.

“Global brands have told us that this is a game changer for managing their brand health,” said Revuze Co-Founder and CEO Ido Ramati. “They are now able to make the kind of key business decisions and utilize data from such varied sources in ways that before were not possible. Through our introduction of AI into this field, brands can now gain an immediate, granular understanding of what customers are saying about their brand, product or competitors, without having to hire teams of experts.”

Revuze is powering global Product Experience Management initiatives for Fortune 500 companies in CPG, food service, leisure and manufacturing. Companies use Revuze for fine tuning customer experience, marketing, product development, business intelligence, business operations, customer service and sales management.

“Revuze is transforming the way that companies track their corporate and product brands through the introduction of AI,” said John Burbank, President, Strategic Initiatives, at Nielsen. “Now, they can easily gain detailed market and consumer intelligence, which enables them to be far more responsive to consumer perception and feedback.”

“The NPD Group is always looking for innovative solutions, and we are impressed with Revuze,” said Steve Coffey, Chief Innovation Officer of The NPD Group. “Revuze is providing a compelling new way to analyze consumer information, making it possible to break down consumer attitudes with much greater accuracy and detail. That’s why we have invested in the company and are taking the extra step of introducing them to our clients.”

About Revuze
Revuze is the AI-powered product experience analyst solution that delivers the most nuanced information about consumers and their needs. The combination of AI, neural networks and machine learning provides immediate, in-depth, ongoing feedback to companies about how their products and services are bought, experienced and viewed. Revuze has created and is widely introducing its technology available for Product Experience Management, and is currently being used by Fortune 500 companies across the CPG, food service, leisure and manufacturing sectors.

Revuze is a venture-backed company with investors that include Nielsen, The NPD Group and TIC Group. The company is headquartered in Netanya, Israel, with U.S. operations in San Francisco, and New York City.

Read More