September 27, 2013
Good news, according to many sources. Google has made changes to its ad-supported, Web search system. A representative write up is “Google Unveils Search Updates for Mobile, New Page Rank Algorithm, and Knowledge Graph Comparisons.”
What could be more useful? Google wants “to help users access information quicker, including new interfaces and features for its iOS and Android apps, along with better integration with the Knowledge Graph.”
Machine-generated comparisons and “knowledge graphs” may prove to be just what searchers like me need to obtain pinpoint results lists. Give the system a whirl. Search for “bwr pwr compare” or “compare pwr bwr” or “pwr versus bwr.” Check out the results. Oh, I guess this did not work. These are nuclear reactor types. Well, maybe in a few days.
When I search for topics like “fluidic self assembly of nanoparticles”, I checked out a knowledge graphs. Oh, I guess this did not work. To see without ads or a fee a “graph”, navigate to www.cluuz.com and try the query. Yes, better. Well, maybe in a short time Google will do the relationship thing in a way that helps me.
The write up asserts:
Google Senior Vice President and software engineer Amit Singhal took the stage afterwards to talk about the company’s future in search. Fifteen years ago, you had to go to a website on a “bulky computer”, turn it on, fire up the dial-up modem, and look up the information and wait for it to be returned. Over time, retrieving answers to questions became possible on the go with the evolution of mobile devices. Singhal says that in today’s age, we’re comfortable with finding information no matter where we are, whenever we want.
My take on “improvements” to Google search include these observations:
- I find it more difficult to locate information today than at any other time in my online experience.
- Results are no longer tied to precision and recall. Results may be hooked to agendas such as advertising revenue. Advertisers pay. The shift to mobile means that the dear Overture approach has to be tweaked. Precision and recall are tossed from the slowing revenue Camry for me.
- Users have zero idea about the accuracy, completeness, or provenance of most search system outputs. Sorry, but I want date and time stamps, information about index freshness, access to content across index silos, and content not distorted by the happy laborers at low cost SEO services.
I could go on, but I won’t. In my lecture at the ISS conference, I told a standing room only crowd, getting online information today in which one has confidence now requires real work. Queries must be passed against multiple search systems. Results lists must be examined carefully. Items of data must be assembled by a human into a coherent fabric. Few have an appetite for this work.
Look at the bright side. Google’s new system makes it easy to find out about Miley Cyrus, rental cars, and pizza. Useful, right?
Stephen E Arnold, September 27, 2013
January 26, 2013
I read “We Are Supposed to Be Truth Tellers.” I think the publication is owned by a large media firm. The point of the write up is that “real news” has a higher aspiration and may deal with facts with a smidgen of opinion.
I am not a journalist. I am a semi retired guy who lives in rural Kentucky. I am not a big fan of downloading and watching television programs. The idea that I would want to record multiple shows, skip commercials, and then feel smarter and more informed as a direct result of those activities baffles me.
Here’s what I understand:
A large company clamped down on a subsidiary’s giving a recording oriented outfit a prize for coming up with a product that allows the couch potato to skip commercials. The fallout from this corporate decision caused a journalist to quit and triggered some internal grousing.
The article addresses these issues, which I admit, are foreign to me. Here’s one of the passages which caught my attention:
CNET reporters need to either be resigning or be reporting this story, or both. On CNET. If someone higher up removes their content then they should republish it on their personal blogs. If they are then fired for that they should sue the company. And either way, other tech sites, including this one, would be more than happy to make them job offers.
I agree I suppose. But what baffles me are these questions:
- In today’s uncertain financial climate, does anyone expect senior management to do more than take steps to minimize risk, reduce costs, and try to keep their jobs? I don’t. The notion that senior management of a media company embraces the feel good methods of Whole Earth or the Dali Lama is out of whack with reality in my opinion.
- In the era of “weaponized information,” pay to play search traffic, and sponsored content from organizations like good old ArnoldIT—what is accurate. What is the reality? What is given spin? I find that when I run a query for “gourmet craft spirit” I get some darned interesting results. Try it. Who are these “gourmet craft spirit” people? Interesting stuff, but what’s news, what’s fact, and what’s marketing? If I cannot tell, how about the average Web surfer who lets online systems predict what the user needs before the user enters a query?
- At a time when those using online to find pizza and paradise, can users discern when a system is sending false content? More importantly, can today’s Fancy Dan intelligence systems from Palantir-likeand i2 Group-like discern “fake” information from “real” information? My experience is that with sufficient resources, these advanced systems can output results which are shaped by crafty humans. Not exactly what the licensees want or know about.
Net net: I am confused about the “facts” of any content object available today and skeptical of smart systems’ outputs. These can be, gentle reader, manipulated. I see articles in the Wall Street Journal which report on wire tapping. Interesting but did not the owner of the newspaper find itself tangled in a wire tapping legal matter? I read about industry trends from consulting firms who highlight the companies who pay to be given the high intensity beam and the rah rah assessments. Is this Big Data baloney sponsored content, a marketing trend, or just the next big thing to generate cash in a time of desperation. I see conference programs which feature firms who pay for platinum sponsorships and then get the keynote, a couple of panels, and a product talk. Heck, after one talk, I get the message about sentiment analysis. Do I need to hear from this sponsor four or five more times. Ah, “real” information? So what’s real?
In today’s digital world, there are many opportunities for humans to exercise self interest. The dust up over the CBS intervention is not surprising to me. The high profile resignation of a real journalist is a heck of a way to get visibility for “ethical” behavior. The subsequent buzz on the Internet, including this blog post, are part of the information game today.
Thank goodness I am sold and in a geographic location without running water, but I have an Internet connection. Such is progress. The ethics stuff, the assumptions of “real” journalists, and the notion of objective, fair information don’t cause much of stir around the wood burning stove at the local grocery.
“Weaponized information” has arrived in some observers’ consciousness. That is a step forward. That insight is coming after the train left the station. Blog posts may not be effective in getting the train to stop, back up, and let the late arrivals board.
Stephen E Arnold, January 26, 2013
December 31, 2012
Searching for information within an enterprise is often not simple, or even fruitful. A recent survey from MindMetre gives us the disturbing truth that over half of knowledge workers admitted they cannot find the information they are looking for using their company’s enterprise search system. The facts are detailed in a white paper from KMWorld titled “Best Practices in Intelligent Search,” which argues that under-management of information, not information overload, is to blame.
An overture to this white paper, “What Are You Looking For? An Overture to ‘Intelligent Search,’” is featured on Enterprise Search Center. The overture states:
“We set out to discuss “enterprise search” and “intelligent search” in this White Paper. And that we shall. But let’s get something straight: Enterprise search is not what you think it is. It is not a single unified piece of software that can magically scour through the dozens of business applications that contain that piece of information our hypothetical guy was looking for. And, much less, it is not a single tool that can seek, discover and deliver an important piece of information from the hundreds and thousands of repositories from which it may emerge.”
This article and corresponding white paper offer a lot of wisdom on enterprise search facilities and Intrafind can be a great resource as organizations seek to make the best of their investments. High-quality research surrounding the capabilities and necessary features of search applications is a focal point of the information retrieval specialists.
Andrea Hayden, December 31, 2012
December 12, 2012
Big data is making the transition from a catchy trend to a serious power in the business world. A flurry of acquisitions involving big data and enterprise search systems are proving that value is being added to big data. According to recent article “Structuring the Unstructured: Why Big Data is Suddenly Interested in Enterprise Search” on CMSWire, this can be attributed to the inclusion of unstructured data (hard-to-predict human content,) into big data analyses, and companies are becoming more and more interested in creating actionable insights from this data set.
The article continues to explain the need to obtain value from unstructured data:
“It is the hands-on application of processes, pragmatism and checksums that produce the most value from unstructured data. A focus on transparency of process creates confidence in data provenance and enables actionable intelligence from unstructured data. That combination of technology and process is what is driving recent acquisitions and what can drive your business to make better, more accurate decisions based on your unstructured big data.”
The whole point of making structured and unstructured data available is so that the right information can drive business decisions. Intrafind makes finding the right data at the right time a bit easier in the age of big data. The company’s software and enterprise search solutions can help you target necessary information from the big data madness and also provide the consultancy services to help you decide what to do with that information.
Andrea Hayden, December 12, 2012
December 3, 2012
Business intelligence (BI) technologies are essential to today’s knowledge workers in the enterprise, providing helpful and necessary decision support software. A recent academic research article addresses the changes and growth in this industry in the past twenty years. The article, “An Overview of Business Intelligence Technology,” published in the August 2011 issue of Communications of the ACM, was written by a team of researchers from Microsoft Research and Hewlett-Packard Lab. According to the article, this rapid growth has been fueled by the declining cost of data storage.
The article states:
“Enterprises today collect data at a finer granularity, which is therefore of much larger volume. Businesses are leveraging their data asset aggressively by deploying and experimenting with more sophisticated data analysis techniques to drive business decisions and deliver new functionality such as personalized offers and services to customers. Today, it is difficult to find a successful enterprise that has not leveraged BI technology for its business.”
This obvious growth accompanies a need to shorten the lag time between acquiring data and decision-making in daily business. Innovations in BI technologies require sophisticated functionality and services. Enterprise search solution Intrafind has a history as a university developed solution that understands the challenges in this industry and the company has created research-based services that solve the problem of making data findable within enterprise environments.
Andrea Hayden, December 3, 2012
November 29, 2012
An impressive showing in second seed round funding–$2.5 million to be exact–was reported by open source file sync and share software company OwnCloud Inc. According to the TechCrunch article titled, “On-Premise Cloud Storage/Sharing Startup, OwnCloud, Closes $2.5M Second Seed Round, Led By General Catalyst Partners,” the round was led by investor General Catalyst Partners as well as new investors.
The Boston-based startup plans to use the money for continued expansion in its enterprise and education customer base as well as support service providers. The article shares:
“‘We’ve been excited about ownCloud since its infancy, and after watching everything they’ve accomplished in a very short time, it validated our belief that they’d developed an exceptional team with an exceptional product,’ said [General Catalyst Managing Director] Bohn in a statement. ‘There’s no one in this increasingly crowded market that can do the things ownCloud does – integrate closely with existing IT, innovate at lightning speed and offer choice of storage locations. With those capabilities already in place differentiating it from the competition, we’re confident that ownCloud will succeed.’”
Serving specific industries and sectors is a great niche for companies concentrating on transformation from storage to deployment. Trusting a solution such as Intrafind, based upon well-formed technology to deploy enterprise solutions that fit the organization, may be a wiser choice in some sectors such as finance and pharmaceuticals.
Andrea Hayden, November 29, 2012
November 26, 2012
Today’s businesses are consistently looking for ways to improve enterprise management and overall operations, so it comes as no surprise that software marketing is targeting those needs. The unfortunate side effect of a good marketing plan can often be a product that does not live up to user expectations. FCW’s article “Does Your Enterprise Search Engine Stink? Here’s Why” reiterates how often times if new software seems too good to be true, it probably is.
Content processing, productivity tools and sorting options are just a few elements of enterprise software that should be evaluated before considering a change:
“Given the predisposition to think of search as pre-formatted to meet our needs, many IT managers and executives believe they can simply purchase, install and operate enterprise search software right out of the box. To a large extent, the leading search software vendors promote this plug-and-play mentality because it is a message customers want to hear. If you are familiar only with Web search as a personal tool, it makes sense to assume that a search engine for your business would operate the same way.”
When relevant results and increased efficiency are the primary concerns a friendly user interface is just an added bonus. Despite marketing hype, quality search requires more than plug and play attention. Full service solutions, expert tagging and feature rich search, such as that provided by Intrafind are a great start for enterprise search solutions.
Jennifer Shockley, November 26 2012
November 2, 2012
The shiny new 7.0 version of Google Search Appliance has been updated for the enterprise, now allowing administrators to add information to the cloud, various social media outlets, and other online storage sites. According to the article “Enterprise Tools Added to Google Search Appliance” on PC Advisor, the upgrade includes a new Entity Recognition feature with auto suggestions for searches as well as a document preview feature.
The article tells us why the need for such an update is necessary for the enterprise:
“IDC analyst David Schubmehl said users would like enterprise searches to be as easy as Web searches, noting that slow searches can hurt productivity. A 2009 IDC study found that the time spent searching for data averaged 8.8 hours per week per employee, at a cost of $14,209 per worker per year.”
We believe Google Enterprise offers some great features, including the option for employees to add their own search results to existing results. However, if secure search and access is an enterprise priority for your corporation, then we would recommend a careful examination before opting for Google Enterprise. A company such as Intrafind offers a secure option for searching structured and unstructured enterprise data.
Andrea Hayden, November 2, 2012
October 29, 2012
Enterprise search platforms are constantly developing and changing to fit the needs of the enterprise, and a new release from DtSearch is attempting to fill some of those needs. We learn about the update in “DtSearch 7.70 Adds Document Filters, Increased Multi-Level Search” on CMSWire, which states the newest version includes enhanced document filters and APIs for OEMs to assist in data management.
The article tells us more about the benefits of the product:
“The search products’ spider can search local/remote content and static/dynamic web content, and it can reach across public/private sites, including support for log-ins and forms-based authentication. More than a terabyte of text can be covered in a single index, including directories, databases, online data and emails, and an unlimited number of indexes can be created and searched. According to dtSearch, that indexed search time is under a second, even across terabytes.”
While the new version does a great job responding to current issues in the enterprise, we believe security in search should be a priority when updating platforms. For a choice that provides high-performance full text search across various content sources while still focusing on access rights and secure search, we recommend Intrafind’s iFinder for enterprise search solutions.
Andrea Hayden, October 29, 2012
July 26, 2012
There is a very enlightening source of reading references to be found in Jeff Huang’s “Best Paper Awards in Computer Science.” He conveniently provided a list of informative papers neatly categorized by area of expertise, like artificial intelligence or human computer interaction.
While scrolling down the list, two interesting papers seemed to jump right out.
The first of which, “Unsupervised Part-of-Speech Tagging with Bilingual Graph-Based Projections,” describes a new approach, as:
“A novel approach for inducing unsupervised part-of-speech taggers for languages that have no labeled training data, but have translated text in a resource-rich language. Our method does not assume any knowledge about the target language (in particular no tagging dictionary is assumed), making it applicable to a wide array of resource-poor languages. We use graph-based label propagation.”
The second paper, “How does search behavior change as search becomes more difficult?” Describes some research on search and their conclusions, with:
“When having difficulty in finding information, users start to formulate more diverse queries, they use advanced operators more, and they spend a longer time on the search result page as compared to the successful tasks. The results complement the existing body of research focusing on successful search strategies.”
Researchers are consistently developing models to predict and understand changes in text entry. Sadly, most of the models fail to account for varying system parameters and the ever changing human factor, nor their evolving relationship.
The latter explains the dumbing of search…but they were interesting reads.
Jennifer Shockley, July 26, 2012