Semantic Machines: A Voice Search Revolution?
January 19, 2016
I read “Newton Startup Scoops Up Talent As It Works to Perfect Artificial Intelligence.” The write up takes an enthusiastic approach to the efforts of a smart software company in the Boston area. I like these types of articles. They remind me of the days when Route 128 was the cat’s pajamas.
I learned that when I talk to my phone, the system is not “smart enough.” I know. Background noise, speaking too quickly, or mumbling are issues with the voice to search thing. Then there is the output. Our test involves asking for the phone number of a person with a Russian name like Kolmogorov in a bus station or a convertible going 40 miles per hour.
The write up points out:
Semantic Machines is currently working on artificial intelligence technology that could do a better job than Siri or other platforms as they interact with users.
There is big money involved; for example, $20 million from the Bainies and other illuminati.
Here’s the angle:
…The idea behind the startup is to develop a “new paradigm” in a field known as conversational computing — essentially improving the way you interact with your phone or computer, whether via voice or text — “much, much closer to the conversational style in the way people talk…”
Worth noting.
Stephen E Arnold, January 19, 2016
Ebay Is Sold Structured Data
January 19, 2016
PayPal and eBay split in 2015 and many people thought it was a poor mistake on eBay’s part. However, eBay has recouped any potential loses by record profits and more than 159 million sellers. Channel Advisor explains that one of the reasons eBay has grown so much is due to its incorporation of structured data and its importance for organic search in the article, “eBay Moves Towards Structured Data-And Why It Matters.”
As an avid eBay buyer and seller, I have been impressed with the new changes in eBay’s demand for structured data. In the past, if you wanted to find anything on eBay you had to go directly to the Web site and dig through results. Sometimes you could find results on Google or another search engine, but these were usually cached auctions. Since the switchover, eBay listings are prominent within Google’s search results. What is even better is how accurate they are!
EBay has turned to structured data as a way to compete with Amazon. While this is beneficial in the long run, it forces sellers to refocus their strategies. The article gives some great tips on how to improve your listings for the best organic search effectiveness. What eBay is demanding now is item specifics so items are placed in the right categories and also helps buyers make more informed decisions. Product identifiers are now very important and mandatory in many categories. These include item specifics such as UPCs, ISBNs, MPNs, GTINs, and more. The goal with all this extra information is to increase visibility in Google and eBay search results.
“In addition to the above benefits, adding identifiers will give you:
- The ability to match your item with a product from the more robust eBay catalogue
- More accurate pricing guidance when you list your items
- Trending price alerts — when your listings are priced lower than the trending price
EBay suggests adding identifiers even if they’re not yet required for your category – doing so will earn you an early competitive edge.”
EBay used to be the one-stop shopping destination online, but Amazon has quickly stolen that title from them. With more detailed listings and visibility in Google, eBay is sure to win back customers.
Whitney Grace, January 19, 2016
Sponsored by ArnoldIT.com, publisher of the CyberOSINT monograph
Many Companies Worldwide Underprepared for Cyber Attacks
January 19, 2016
A recent survey from KPMG Capital suggests that only about half the world’s CEOs feel their companies are “fully prepared” to counter a cyber breach in the next three years. One notable exception: businesses in the U.S., where about ninety percent of CEOs feel their companies are ready to fend off hackers. We are not surprised that KPMG is gathering information on in the subject, since it recently took an equity stake in cyber-intelligence firm Norse Corp.
KPMG Australia comments on the survey’s results in its post, “Cyber Security: A Failure of Imagination.” The write-up relates:
“According to the 2015 KPMG CEO Outlook Study [PDF] of more than 1,200 CEOs, one out of five indicated that information security is the risk they are most concerned about. ‘Collectively we sleepwalked into a position of vulnerability when it comes to cyber,’ said Malcolm Marshall, Global Head of Cyber Security at KPMG. ‘This combination of lack of preparedness and concern, from those organizations that are among the best equipped to deal with risks of this magnitude, clearly illustrates cyber security challenges remain severely unaddressed.’”
A lack of skilled cyber-security workers seems to be a large part of the problem, particularly ones who also have management or social-science skills. However, we’re told the root cause here is the “failure to imagine” what hackers can do and might try before they’ve tried it. Clearly, many executives would do well to get themselves up to speed on the subject, before their companies fall victim.
Cynthia Murrell, January 19, 2016
Sponsored by ArnoldIT.com, publisher of the CyberOSINT monograph
Ah, Ha. Search May Be Dying
January 18, 2016
I did not know that. I am delighted to have wisdom available from a blog focusing on search engine optimization.
Dear, old search. Are you really dying? I thought you had pulled off one of the Carlos Castaneda transmogrifications into augmented intelligence, customer support services, analytics, and my favorite Big Data. If Vivisimo can pull this off at IBM, almost any company with information access capabilities can wake up as a metasearch company and go to bed as a Big Data champ with the four Vs dancing in one’s dreams.
The write up points out:
Recent years have revealed a worrisome trend (for Google anyway) — search engine use overall has declined from 90 percent in 2009 to 86 percent in 2014. This might not seem like much of a downward trend, but if you consider that overall global Internet use has increased by 67 percent in the same period, that’s a lot of Internet users who aren’t turning to search.
The article represents this wonderful Pew Research chart:
But the section which tickled my Alphabet Google fancy was this passage:
There’s no denying that Google is the most complex searchable database on the Internet. It offers billions of results and is constantly innovating new ways to determine your search needs. However it would seem that Google’s impressive scope is the very thing that is sending people to apps and other websites to find the information they need. People want results that are personalized for them, while Google is busy trying to be everything for everyone. There are simply too many relevant results in Google’s database to match the personalization capabilities of apps and websites. That’s why apps are increasingly being used as research channels, especially among teens, who are 30 percent more likely to use them for search.
The inescapable conclusion seems to be that search is a goner.
I don’t agree, but that’s not germane to the SEO mavens who stand ready to serve customers eager for clicks, app downloads, and lots of SEO goodness.
At least the Mirror comes at the Alphabet Google thing with a bit of creativity. See, for example, “Google Slammed over Refusal to Advertise Plus Size Fashion to Curvy Consumers.” Now that’s real hard hitting evidence for the argument that search is on a down ward trajectory.
Stephen E Arnold, January 18, 2016
Stephen E Arnold
Surprise. News Corp. Thinks Probe of Google Is Okay
January 18, 2016
I love it when outfits which are engaged in real journalism comment about the Alphabet Google thing. Newspapers never had a monopolistic thought in their history. The newspaper wars are a myth. Yellow journalism was a result of color blind folks misunderstanding “real” data gathering and dissemination.
To learn more about News Corp.’s view of a Google anti trust probe, navigate to “News Corp. CEO Says Antitrust Probe of Google Could Be Warranted.” Yep, woulda, coulda, shoulda.
I learned:
Citing a lack of competition in the search market, especially in Europe, News Corp. CEO Robert Thomson called Monday for closer scrutiny of Google’s business by regulators.
I wonder if the News Corp. digital wizards is aware of:
- Exalead Search at https://www.exalead.com/search/
- Gibiru at https://anonymous-gibiru.com/
- Qwant at https://www.qwant.com/
- Silobreaker at http://www.silobreaker.com/
- Unbubble at https://www.unbubble.eu/
- Yandex at https://www.yandex.com/
There are some other services as well, but from the point of view of New Corp., they are probably irrelevant.
I assume that alternatives are not important to executives representing a company with an interesting track record in management and information gathering.
I wonder why folks are using the Alphabet Google thing? Thoughts, gentle reader?
Stephen E Arnold, January 18, 2016
AOL: A New Do for the Year of the Monkey
January 18, 2016
I read “AOL’s Identity Crisis: The Company May Ditch the AOL Brand.” I remember the flood of discs. I remember the hidden files thoughtfully placed on my hard drive when I installed America Online. I remember the Xoogler who bought his own local publishing outfit to reinvigorate AOL and, of course, his own local America Online. I remember Verizon buying AOL because, well, it could.
According to the write up:
one of AOL’s biggest priorities for the new year is figuring out its brand and investing in it, even if that means saying goodbye to the name “AOL” in favor of launching something completely new.
I learned that
Mark Ritson, a leading brand expert and marketing consultant, tells Business Insider that he also thinks the messy corporate brand definitely needs a clean up. AOL is tricky, he says, because it has very strong brand awareness, but that its image “has an unpalatable mix of being seen to be out of date and a business failure.”
No matter what name Verizon chooses, AOL will always evoke fond memories for me. The dial up modem, the chat groups, the fantastical email services.
So many memories. What ever happened to that Xoogler’s local news idea? Ah, it did not work out.
I look forward to the Yahooligans following AOL’s trajectory. Two new branding opportunities for the marketing consultants.
It is the year of the monkey too. Love those creatures.
Stephen E Arnold, January 18, 2016
DtSearch in the Cloud
January 18, 2016
Enterprise- and developer-search firm dtSearch now offers a platform for the cloud. PR.com informs us, “New .NET Solution Uses dtSearch with Microsoft Azure Files and RemoteApp.” The solution allows users to run the dtSearch engine entirely online with Microsoft Azure, ensuring their security with Microsoft’s RemoteApp. The press release elaborates:
“The solution enables cloud operation of all dtSearch components, leveraging Microsoft’s new Azure Files feature for dtSearch index storage. Searching (including all 25+ dtSearch search options) runs via Microsoft’s RemoteApp. Using RemoteApp gives the search component the ‘look and feel’ of a native application running under Windows, Android, iOS or OS/X. Developers using dtSearch’s core developer product, the dtSearch Engine, can find the solution on CodeProject, including complete Visual Studio 2015 .NET sample code.”
See the thorough write-up for many details about the product, including supported formats, search and classification options, and their terabyte indexer. We note, for example, the capacities for concurrent, multithreaded search and for federated searches with their dtSearch Spider.
Founded in 1991, dtSearch supplies search software to firms in several fields and to numerous government agencies around the world. The company also makes its products available for incorporation into other commercial applications. dtSearch has distributors worldwide, and is headquartered in Bethesda, Maryland.
Cynthia Murrell, January 18, 2016
Sponsored by ArnoldIT.com, publisher of the CyberOSINT monograph
Machine Learning Hindsight
January 18, 2016
Have you ever found yourself saying, “If I only knew then, what I know now”? It is a moment we all experience, but instead of stewing over our past mistakes it is better to share the lessons we’ve learned with others. Data scientist Peadar Coyle learned some valuable lessons when he first started working with machine learning. He discusses three main things he learned in the article, “Three Things I Wish I Knew Earlier About Machine Learning.”
Here are the three items he wishes he knew then about machine learning, but know now:
- “Getting models into production is a lot more than just micro services
- Feature selection and feature extraction are really hard to learn from a book
- The evaluation phase is really important”
Developing models is an easy step, but putting them in production is difficult. There are many major steps that need attending to and doing all of the little jobs isn’t feasible on huge projects. Peadar recommends outsourcing when you can. Books and online information are good reference tools, but when they cannot be applied to actual situations the knowledge is useless. Paedar learned that real world experience has no comparison. When it comes to testing, it is a very important thing. Very much as real world experience is invaluable, so is the evaluation. Life does not hand perfect datasets for experimentation and testing different situations will better evaluate the model.
Paedar’s advice applies to machine learning, but it applies more to life in general.
Whitney Grace, January 18, 2016
Sponsored by ArnoldIT.com, publisher of the CyberOSINT monograph
AI News: Bad Links Resulted in My Finding This Smart Software Post
January 17, 2016
I followed some bad links and after a bit of poking around, I arrived at “A Tour of Machine Learning Algorithms.” The write up is more than two years old, but it does have some useful information to feed the hungry minds of the smart software crowd.
The write up explains machine learning algorithms group by learning style and by similarity. The brief explanations are helpful.
The major omission is that none of the algorithms apparently has any notable flaws or vulnerabilities. In our research, we found that the algorithms we tested could be manipulated by creating data which threw the method off the scent for accuracy.
One example is the flaw of a properly “trained” Bayesian system. By loading content outside the original training set’s scope, the Bayesian system happily generated off point results. This means that directed content injection such as shaped or weaponized blog posts or social media message flooding can make the systems wander.
Good news for those who want to manipulate smart software. Bad news for the decision makers who “assume’ the Bayesian system’s outputs remain on the mark.
Useful to know in my opinion. Descriptions of algorithms have to include their less sunny side, don’t you think?
Stephen E Arnold, January 17, 2016
Watson Weekly: Entrepreneurs, Watson Will Help You
January 17, 2016
I think that folks should help themselves. IBM is a giant struggling to generate revenues. The company’s approach to innovation is to buy companies which duplicate functions the firm already has. I assume that IBM’s senior executives ask Watson what to do, and s/he suggests, “Buy more companies like those you already own, pilgrim.”
I read “How IBM Watson Can Help Developers and Entrepreneurs with its Services”. I learned that
Technology giant IBM has been helping people from all walks of life,, especially developers who are prototyping and building Cloud-based cognitive computing applications, through Watson since its launch in 2011.
Okay, going on six years of helping.
Even better, IBM is now
allowing a growing ecosystem of developers, advertising professionals, businessmen, bankers, students, entrepreneurs, tech enthusiasts and people from various professions to use this advanced cognitive computing platform.
So who is being helped? IBM or the developers? The write up clears this up for me:
Kambhatla [An IBM cognitive computing wizard] described Watson Developer Cloud as one stop shop for developers. The developers can also bring their own data to many of the services, to help train Watson about a specific domain or industry and customize API models to speed up and simplify the development of cognitive apps. For instance, Watson can be used to help play relevant advertisements anywhere, be it shopping complex, roadside or amusement park based on image of people in the vicinity, as it can guess approximate age, gender, etc.
Okay, but it has been five years. Where’s the revenue? Let’s ask Watson. Okay, but I can’t find a public general purpose IBM Web site to allow that to happen.
Gee, maybe Watson’s advice is too darned good to make available to the addled goose? The PR folks obviously have access. The senior managers have access. Kambhatla has access. Why the shroud of unknowing, dear Big Blue?
Stephen E Arnold, January 17, 2016
-
- Subscribe to Beyond Search
Feature archive
News archive
-
