IBM Watson Performance: Just an IBM Issue?
September 6, 2017
I read “IBM Pitched its Watson Supercomputer As a Revolution in Cancer Care. It’s Nowhere Close.” Here in Harrod’s Creek, doubts about IBM Watson are ever present. It was with some surprise that we learned:
But three years after IBM began selling Watson to recommend the best cancer treatments to doctors around the world, a STAT investigation has found that the supercomputer isn’t living up to the lofty expectations IBM created for it. It is still struggling with the basic step of learning about different forms of cancer. Only a few dozen hospitals have adopted the system, which is a long way from IBM’s goal of establishing dominance in a multibillion-dollar market. And at foreign hospitals, physicians complained its advice is biased toward American patients and methods of care.
The write up beats on the lame horse named Big Blue. I would wager that the horse does not like being whipped one bit. The write up ignores a problem shared by many “smart” software systems. Yep, even those from the wizards at Amazon, Facebook, Google, and Microsoft. That means there are many more stories to investigate and recount.
But I want more of the “why.” I have some hypotheses; for example:
Smart systems have to figure out information. Now on the surface, it seems as if Big Data can provide as much input as necessary. But that is a bit of a problem too. Information in its various forms is not immediately usable in its varied forms. Figuring out what information to use and then getting that information into a form which the smart software can process is expensive. The processes involved are also time consuming. Smart software needs nannies, and nannies which know their stuff. If you have ever tried to hire a nanny who fits into a specific family’s inner workings, you know that the finding of the “right” nanny is a complicated job in itself.
Let’s stop. I have not tackled the mechanism for getting smart software to “understand” what humans mean with their utterances. These outputs, by the way, are in the form of audio, video, and text. To get smart software to comprehend intent and then figure out what specific item of tagged information is needed to deal with that intent is a complex problem too.
IBM Watson, like other outfits trying to generate revenue by surfing a trend, has been tossed off its wave rider by a very large rogue swell: Riffing on a magic system is a lot easier than making that smart software do useful work in a real world environment.
Enterprise search vendors fell victim to this mismatch between verbiage and actually performing in dynamic conditions.
Wipe out. (I hear the Safaris’ “Wipe Out” in my mind. If you don’t know the song, click here.)
IBM Watson seems to be the victim of its own over inflated assertions.
My wish is for investigative reports to focus on case analyses. These articles can then discuss the reasons for user dissatisfaction, cost overruns, contract abandonments, and terminations (staff overhauls).
I want to know what specific subsystems and technical methods failed or cost so much that the customers bailed out.
As the write up points out:
But like a medical student, Watson is just learning to perform in the real world.
Human utterances and smart software. A work in progress but not for the tireless marketers and sales professionals who want to close a deal, pay the bills, and buy the new Apple phone.
Stephen E Arnold, September 6, 2017
Are Vendors of Enterprise Search Distracted?
September 6, 2017
I read “To Have Good Ideas, Remember to Get Bored.” I noted this assertion in the write up:
the temptation of constant podcast listening, phone fiddling, and TV watching takes over. The fight to maintain some boredom never ends.
The idea is that distraction kills boredom. Without boredom, “people” do not get good ideas. Ergo when one notes lots of bad ideas, that may be the signal that distraction undermines innovative thinking.
I am not certain the statements in the write up and the accompanying TED talk (which bored me, by the way) are applicable across a population sleeted at random in Rwanda or rural Kentucky, but let’s assume the idea has value.
I look at enterprise search and I see the same old perpetual motion machines: Semantics, metatagging, context, yada yada.
Perhaps those involved in enterprise search system development are manifesting their distractedness. Instead of putting down the mobile and performing myriad displacement activities, are enterprise search system developers fresh out of ideas.
Something’s wrong. Analysts find search just peachy when relying on SAP, IBM Watson, Fabasoft, and the other systems available today.
I know I am bored, and I would postulate that those involved in next generation information access systems may want to cultivate a bit of boredom as well. Innovation may come about.
Example: As I was thinking about today’s me-to enterprise search systems, I was bored. I decided to begin work on a new book in my cyber intelligence series. How does eDiscovery for Investigators sound?
Boring?
Stephen E Arnold, September 6, 2017
Blockchain Quote to Note: The Value of Big Data as an Efficient Error Reducer
September 6, 2017
I read “Blockchains for Artificial Intelligence: From Decentralized Model Exchanges to Model Audit Trails.” The foundation of the write up is that blockchain technology can be used to bring more control to data and models. The idea is an interesting one. I spotted a passage tucked into the lower 20 percent of the article which I judged to be a quote to note. Here’s the passage I highlighted:
as you added more data — not just a bit more data but orders of magnitude more data — and kept the algorithms the same, then the error rates kept going down, by a lot. By the time the datasets were three orders of magnitude larger, error was less than 5%. In many domains, there’s a world of difference between 18% and 5%, because only the latter is good enough for real-world application. Moreover, the best-performing algorithms were the simplest; and the worst algorithm was the fanciest. Boring old perceptrons from the 1950s were beating state-of-the-art techniques.
Bayesian methods date from the 18th century and work well. Despite LaPlacian and Markovian bolt ons, the drift problem bedevils some implementations. The solution? Pump in more training data, and the centuries old techniques work like a jazzed millennial with a bundle of venture money.
Care to name a large online outfit which may find this an idea worth nudging forward? I don’t think it will be Verizon Oath or Tronc.
Stephen E Arnold, September 6, 2017
Insight Engines Are the next Enterprise Upgrade
September 6, 2017
When one buzzword loses its, marketing teams do their best to create the next term to stay on top of their competition. When it comes to search, the newest buzzword appears to be “insight engine.” Mindbreeze is top-selling insight engine, according to their Web site and the recent blog post, “The Global Insight Engine Market: European Solution Scores Top Position.” The post makes a poignant point that quickly retrieving answers to complicated problems is a necessity, but regular enterprise search engines cannot crawl unstructured information.
While insight engines are the next buzzword and also the next generation of enterprise search engines, but what exactly do their do?
This is where so-called insight engines come into play. They interpret unstructured and structured data using semantic analysis, and prepare it for further use. Search results are improved and returned in a structured format. Of course, insight engines don’t just process unstructured information, but also all other existing company information. The connection to the individual data sources is made through so-called connectors. Another feature of insight engines is that search queries can be formulated in natural language. The intelligent tools interpret the query and provide the relevant corresponding search results.
Gartner recently ranked the global insight engines market (they have their own market from other search engines?) and Mindbreeze ranks at the top of all the engines in the “challenger” category. What makes this a headliner is that Mindbreeze competed against IBM and HP. Mindbreeze then brags about their features: less than 90 days to integrate into a system, more out-of-the-box solutions for data connectors than other vendors, and Mindbreeze is more popular now since Google withdrew from the market.
Since this was published on Mindbreeze’s own blog, of course, it is a publicity piece. In an objective test, how would Mindbreeze compete against Europe’s other engine, Elasticsearch?
Whitney Grace, September 6, 2017
Facebook Unapologetic About Spy Tool
September 6, 2017
At what point does a company or industry hold too much power? That is exactly what a recent TNW article examined. According to the site, Facebook has unleashed an early spying tool to identify and then eradicate competition. Many examples of how Facebook has done this in the past, stealing such features as Stories or upcoming Bonfire, from start-ups, are listed as proof of the growing power the social media giant possesses.
But it doesn’t stop there.
Amazon, Microsoft, Apple, and others all wield the same sort of power over smaller competitors. While the power shift isn’t revolutionary at its surface — offline businesses held the same sort of power for decades, and some still do — it’s the speed at which online companies grow, becoming ever-more-powerful, that makes it worth taking notice of.
With just a handful of companies (Google, Facebook, Apple primarily) holding so much revenue power in the global economy, it is important for us not to just gloss over these practices. What the future will hold for new companies with bold, new ideas is daunting, at the very least.
Catherine Lamsfuss, September 6, 2017
Another Tale of Googzilla: The Browser Discontinuity
September 5, 2017
I read “My Friends at Google: It Is Time to Return to Not Being Evil.” The write up is a gentle, polite first person narrative about one browsers brush with the Google. Who knows if the narrative reflects the GOOG’s side of the story. I found these statements interesting. You, gentle reader, will have to consult the original and make your own decision.
ITEM 1
Google increased their proximity with the Mozilla foundation. They [Google] also introduced new services such as Google Docs. These services were great, gained quick popularity, but also exposed the darker side of Google. Not only were these services made to be incompatible with Opera, but also encouraged users to switch their browsers.
ITEM 2
Recently, our Google AdWords campaigns were suspended without warning [for the Vivaldi browser]
ITEM 3
Two days after my thoughts were published in an article by Wired, we found out that all the campaigns under our [Vivaldi’s] Google AdWords account were suspended – without prior warning…When we reached out to Google to resolve the issue, we got a clarification masqueraded in the form of vague terms and conditions… In exchange for being reinstated in Google’s ad network, their in-house specialists dictated how we should arrange content on our own website and how we should communicate information to our users.
ITEM 4
A monopoly both in search and advertising, Google, unfortunately, shows that they are not able to resist the misuse of power.
Intriguing if the information is accurate. Is the Google not the lovable, friendly outfit we love so much in Harrods Creek? Nah, as the write up explains:
After almost three months of back-and-forth, the suspension to our account has been lifted, but only when we bent to their requirements.
No problemo, right?
Stephen E Arnold, September 5, 2017
.
Do You See How Search Will Change?
September 5, 2017
Vocal-activated search is a convenient, hands-free way to quickly retrieve information. A number of people who use some form of vocal search, either using a smart speaker or a digital assistant. Scott Monty reports that the voice-activated speaker market has increased by 130% in the article, “Is The Future Of AI-Powered Search Oral Or Visual?” Amazon controls 70% of the smart speaker market, while Google has 23%.
Voice activated search has its perks, but it does not always prove to be the most useful. The problem with voice-activated search is that it does not allow a lot of options:
But here’s the current challenge with voice-activated systems: there’s no menu. There’s no dropdown of options. There’s no visual cue to help you give you a sense of what you can ask the system. Oh sure, you can ask what your query options are, but the voice will simply read back to you what your options are.
Monty points out that humans have been a visually-driven culture for thousands of years, ever since written language was invented. Amazon and Google are already working on projects that combine visual aspects with voice-driven capabilities. Amazon has the Echo Snow that has the same functionality as the regular Echos, except it has a screen. Google is developing the Google Lens; think Google Glasses except not as obtrusive. It can use visual search to augment reality. The main differences between the two companies still leave a big gap between them: Amazon sells stuff, Google finds information.
But here’s the current challenge with voice-activated systems: there’s no menu. There’s no dropdown of options. There’s no visual cue to help you give you a sense of what you can ask the system. Oh sure, you can ask what your query options are, but the voice will simply read back to you what your options are.
Google still remains on top, but Amazon could develop an ecommerce version of the Google Lens. Or would it be easier if the two somehow collaborated on a project to conquer shopping and search?
Whitney Grace, September 5, 2017
Audioburst Tackling Search in an Increasing Audio World
September 5, 2017
With the advent of speech recognition technology our Smart world is slowly becoming more voice activated rather than text based. One company, Audioburst, is hoping to cash in on this trend with a new way to search focusing on audio. A recent TechCrunch article examines the need for such technology and how Audioburst is going about accomplishing the task by utilizing natural language processing and speech recognition technology to identify and organize audio data.
It…doesn’t only match users’ search queries to those exact same words when spoken, either. For example, it knows that someone speaking about the “president” in a program about U.S. politics was referring to “Donald Trump,” even if they didn’t use his name. The audio content is then tagged and organized in a way that computers understand, making it searchable…This allows its search engine to not just point you to a program or show where a topic was discussed, but the specific segment within that show where that discussion took place. (If you choose, you can then listen to the full show, as the content is linked to the source.)
This technology will allow users to never need the physical phone or tablet to conduct searches. Audioburst is hoping to begin working with car manufacturers soon to bring truly hands-free search to consumers.
Catherine Lamsfuss, September 5, 2017
Google: Is the Company the Avis to Amazon Hertz?
September 4, 2017
Holiday time. You can think about the information in “Tech Companies Spend More on R&D Than Any Other Companies in the U.S.” My recollection is that this type of list was once the backyard for Business Week. Now it is the talk of a new “real” journalism outfit. The overall listing does not seem in or out of whack.
Wait, wait, do tell me the answer to these questions:
- Does the Google number include today’s “20 percent free time”?
- Does the Google number take into account the research activities buried within certain operating groups; for example, the merry band of “let’s make Google Plus great”?
- Does the Google number factor in the research conducted when eager Googlers team up with academics to probe the frontiers of science and technology?
If the answer to the questions is “Here are the breakouts?”—what’s the breakout for the Bezos machines?
Without granularity, my hunch is that both Amazon’s and Google’s R&D expenditures are not comprehensive.
In my 2003 The Google Legacy, I did some basic math about the dollar contribution of the then-20 percent time “value” for R&D. My recollection was that it was a big number.
Just a game show quiz question for this holiday, “How big?”
Stephen E Arnold, September 4, 2017
PayPal and eBay Used to Smuggle Funds, According to FBI
September 4, 2017
Online is an exciting place. Now, eBay and PayPal appear to have unwittingly hosted the transfer of terrorist funding, we learn from an article at The Next Web titled, “FBI Says ISIS Smuggled Funds to US Using eBay and PayPal.” Citing reporting by The Wall Street Journal, writer Rachel Kaser reveals:
An FBI affidavit alleges that the Islamic State used everyone’s favorite digital auction house to transfer cash to one of its US-based agents. The agent was disguising himself as a printer salesman — he’d pretend to sell a printer, only to receive payment from IS via eBay and PayPal. Supposedly, it was all part of a network operated by the late Siful Sujan, who was at one point a director of ISIS’s computer operations. The FBI document claims he’s just one of a network of agents stretching from the UK to Bangladesh. It doesn’t say whether they all used eBay to fund their schemes. The suspect in this case apparently used the money he received from the printer sales to buy a laptop, a cellphone, and a VPN.
An eBay spokesperson emphasized their company’s “zero tolerance” for criminal activity on their platform. The company is cooperating with authorities, and the alleged transferor of terrorist funds is awaiting trial.
Cynthia Murrell, September 4, 2017
-
- Subscribe to Beyond Search
Feature archive
News archive
-
