Yandex Outpaces Microsoft in Search Traffic
February 27, 2013
We knew Yandex was one to keep an eye on. Now, Search Engine Watch announces, “Yandex Just Passed Bing to Become 4thLargest Global Search Engine.” According to comScore‘s recent qSearch report analyzing traffic from last November and December, Yandex processed 4.844 billion queries to Microsoft’s 4.477 billion. This despite the fact that the report lumps together traffic from Bing with that of other Microsoft properties like MSN and Windows Live. (Google, of course, is still way, way, way ahead with 114.73 billion search queries.)
Writer Michael Bonfils notes that we cannot be sure what drives this Yandex lead. He observes, though, that the pool of Russian users is still growing and evolving, while the Western markets where Microsoft processes the most traffic seem to be approaching saturation. The article concludes:
“You can speculate all day about what’s happening here. Do Russians just search more? Are Russians searching more because they don’t like the results? Are they gaining market share in countries like Turkey or the Ukraine? Who wins with unique users?
“Regardless of all that, I wasn’t expecting to see Yandex, which doesn’t have nearly the marketing budgets of Microsoft, surpass them in global search queries by the end of 2012. Nothing better than seeing incredibly talented underdogs race past one of the biggies.”
Though we find this development less of a surprise, we join Bonfils in cheering for the underdog. What will the future bring for the formidable Yandex?
Cynthia Murrell, February 27, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
Hakia to Revolutionize Online Shopping Experience
February 25, 2013
Canadian start-up Flow, is teaming up with Hakia to provide semantic search capabilities within a closed platform and take online shopping into a new realm in terms of product specificity and search capability.
Until now e-commerce sites have followed roughly the same pattern; Ebay and Amazon, both powerhouses in the online shopping experience introduced a platform for “every product” and have made a lot of money in the last decade.
The article from Silicon Angle, “Flow Adds Semantic Search from Hakia to Revolutionize E-Commerce,” lays out how Flow and Hakia are creating a different way to search for the products you are looking for, without having to wade through all the extraneous mumbo jumbo. This new partnership hopes to do is to create a social flow and eliminate the sixth degree of separation.
“eCommerce as we know it is pretty entrenched, but social commerce is slowly emerging to challenge the status quo. It’s a concept that’s evolved from what are probably the two biggest phenomena on the web – online shopping and social media. And it’s a natural evolution too, as it only makes sense for marketers to connect with their customers to better understand their needs and position themselves as the ones to provide it.”
Facebook is probably the biggest example of the social marketplace at the moment. Facebook isn’t a shopping powerhouse because it has no search structure. Utilizing semantic search is going to create a kind of exclusive marketplace that hopes to promote less cutthroat competition; since users will be finding exact matches for their searches there’s no competition for most hits in order to remain at the top.
But can the Flow/Hakia partnership really pull through with those kinds of promises? It seems like a pretty tall order to fill. Functionality and no middlemen sound like a dream come true to eCommerce consumers, but the proof is in the pudding.
Leslie Radcliff, February 25, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
Twitter Expands Search Reservoir
February 25, 2013
Twitter, now with more tweets! I suppose that’s a real plus for some. ComputerWorld announces, “Twitter Search to Show Tweets More Than a Week Old.” Writer Jeremy Kirk explains:
“Twitter is modifying its search engine to include tweets more than a week old, a move it said will help users uncover better content.
“Over the next few days, searches will return ‘a fairly small percentage of total tweets ever sent’ but that will increase over time, wrote Paul Burstein, an engineer who works on Twitter’s search infrastructure, on a company blog.
“‘We look at a variety of types of engagement, like favorites, retweets and clicks, to determine which tweets to show,’ Burstein wrote. ‘We’ll be steadily increasing this percentage over time, and ultimately, aim to surface the best content for your query.'”
I suppose a wider search results field is better, whatever the platform. The expansion was announced alongside Twitter‘s updates to its search function in its iOS and Android mobile apps. These apps now return tweets, photos, and people in a single results stream rather than separate tabs. One other change saves users a step by letting them go directly to a linked Web site without opening the corresponding tweet. Ah, the relentless march of progress—now saving us a few seconds at a time.
Cynthia Murrell, February 25, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
Sinequa Highlighted by Yandex
February 22, 2013
I noticed this morning that Sinequa, a vendor which provides a unified information access solution has been moved to the top of the Yandex results list. I was poking around the Yandex system to see how the Google challenger was handling some European analytics, content processing, business intelligence, and search vendors.
On a previous test, Yandex displayed a link to a site offering translations of Latin phrases. (Sinequa, as you may recall from your school days, can be translated as “an essential component or element.” I had one Latin teacher suggest that sinequa indicated “none higher.” Yet another of the specialists who with whom I studied boiled the connotation down to “excellence.”)
CrunchBase describes the company this way:
Sinequa helps companies and organizations to cope with the data explosion and enterprise transformation. Sinequa’s unique value proposition is to provide an out-of-the-box enterprise search solution leveraging all enterprise content, resulting in significant savings for large organizations. Sinequa’s customers are some of the most innovative companies in the world including Siemens, Mercer, EADS, Saint-Gobain, SalesForce.com, Bouygues, SFR, Atos Origin, Loreal, LCF Rothschild, Credit Agricole, the French Ministry of Defense, Groupe Figaro.
You can obtain more information about the company at www.sinequa.com. As an aside, I find the Yandex results increasingly useful. Check the system out at www.yandex.com.
Stephen E Arnold, February 22, 2013
Twitter Redesigns Search Feature
February 22, 2013
Twitter’s blog post, “Search and Discover Improvements: Get More Great Content Faster,” describes updates to the service’s Android and iOS apps and to its mobile-tailored Web address . The primary change, as revealed by product management director Esteban Kozak, is the implementation of separate tabs with their own content streams.
There four of these distinct streams— Discover, Search, Connect, and Links. It is the first two that we find most interesting. The write-up specifies:
“Discover: Now all the content in Discover — Tweets, Activity, Trends and suggestions of accounts to follow — appears in a single stream, on both iPhone and Android. You can also dive into Activity and Trends from new previews at the top of the Discover tab.
“Search: Search results now surface the most relevant mix of Tweets, photos, and accounts, all in one stream (similar to the stream in Discover). We’ve also added a new search button to Twitter for iPhone, letting you search from anywhere within the app. (This button was already available in the Android and iPad apps.) Look for the magnifying glass icon next to the button you use to compose a Tweet.”
Making search and discovery easier to find and use is a worthy goal, and usually fairly straightforward to implement. The quality of search results, it should be remembered, is another matter entirely. The post mentions that information on new developments can always be found within Twitter’s entries at the App Store and Google Play.
Cynthia Murrell, February 22, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
Oracle Text Workaround for Stop Words List
February 22, 2013
We’ve come across a discussion about Oracle Text at StackOverflow, “Oracle Text Search Doesn’t Work on Some Words.” Essentially, some words cannot initially be indexed, and the fix is to go in and remove those words from the stop words list. Interesting.
The question-and-answer site for programmers received this query:
“I am using Oracle’ Text Search for my project. I created a ctxsys.context index on my column and inserted one entry ‘Would you like some wine???’. I executed the query ‘select guid, text, score(10) from triplet where contains (text, ‘Would’, 10) > 0′; it gave me no results. Querying ‘you’ and ‘some’ also return zero results. Only ‘like’ and ‘wine’ matches the record. Does Oracle consider you, would, some as stop words?? How can I let Oracle match these words? Thank you.”
The top response reveals:
“I found that the query’s output is perfect according to the stop word lists that is in the oracle.
those words can be found in the ctxsys package, and you could query for the stoplist and the stop words using “SELECT * FROM CTX_STOPLISTS; SELECT * FROM ctx_stopwords;” and yes, the oracle consider ‘you’, ‘would’ in your query as stop words.”
The solution—remove the offending stop words with the command, “GRANT EXECUTE ON CTXSYS.CTX_DDL to you” followed by the desired procedure. See the link for an example.
Cynthia Murrell, February 22, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
Bad News for Old Style Web Traffic
February 21, 2013
Who knows how universal the information in “Why No One Is Talking about Yahoo’s – or Anyone Else’s – New Homepage.” I find the idea that traditional Web traffic is changing quite interesting. Here’s the snippet I tucked into my traditional 4×6 inch note card box which I kept since my high school debating days:
Like other titans of another tech age, Yahoo is facing an existential threat against which it may be defenseless: People just don’t surf the web the way they used to. It is now the rule, rather than the exception, to share links over Facebook, Twitter and “dark social“ (e.g., email or text messages), which means that most people are arriving on pages buried deep within websites, and may never go near the homepage.
There are some interesting implications for search and retrieval. How does one get found? Good question. Answer: Augmentext.
Stephen E Arnold, February 21, 2013
Oracle Endeca
February 21, 2013
It appears that Oracle is effectively marketing its search acquisition, Endeca. The company’s blog posts a piece on its business intelligence features in, “Transforming Workforce Management with Oracle Endeca Information Discovery.” Meanwhile, the Quest 4 ATG blog shares some migration tips for the commerce platform in its piece, “Stepping into Oracle Endeca Commerce 3.1.1.”
Oracle’s BI post asserts that Human Resources is where the action is for companies right now, but that the challenges of accessing unstructured data can stand in the way of making the best decisions in that area. Their Endeca Information Discovery, they say, can help by:
- Understanding the voice and sentiment of the workforce. By providing the ability to use natural-language queries to derive insights from information from all sources, Oracle Endeca Information Discovery provides insight into the effectiveness of strategic and tactical HR decisions. . . .
- Identifying the root causes behind events and exceptions. . . . HR organizations are empowered to determine the reasons why things happen and respond to findings without direct IT involvement, saving time and cost.
- Supercharging existing business analytics investments. With the unscripted exploration of workforce data and full-featured search, navigation & interactive analytics, HR organizations can leverage, improve, and extend existing investments in workforce analytics and HCM applications.
The post at Quest 4 ATG is more of a practical matter. It documents that blogger’s migrations from Endeca Commerce version 3.1.0 to version 3.1.1. Clearly described in outline form, with illustrations, these instructions are good to note (and bookmark) if you, too, need to migrate.
Cynthia Murrell, February 21, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
Upcoming Webinar on Optimized Search
February 20, 2013
LucidWorks is a major contributor to open source technology, training, and development. In addition to supporting the development of open source projects, LucidWorks also offers training on implementation. One of the latest training opportunities to be made available comes in the form of a webinar. Read the details in, “Webinar: Intersection of Search, Analytics and Big Data.”
The registration information begins:
“There is a perfect storm brewing: organizations are amassing large volumes of multi-structured data, yet there are few cost-effective ways to analyze this data without data scientist, who seem to be as elusive as a white unicorn. Although there are higher level languages on the horizon, like Hive and Pig, that promise to give mere mortals access to Hadoop-bound data, there is an easier approach that some may consider old school: search technology.”
And LucidWorks happens to specialize in search technology. In the webinar you can plan to learn:
- “The value of using search to perform analytics
- The marriage of search and big data
- The technology that powers search-enabled analytics
- Use cases for search-enabled analytics”
The presenter is Wayne Eckerson, the director of research at TechTarget. Take advantage of this professional training opportunity from LucidWorks and learn more about how search technology will not only make sense of unstructured data, but can actually turn it into a powerful analytical advantage.
Emily Rae Aldridge, February 20, 2013
Sponsored by ArnoldIT.com, developer of Beyond Search
Creating Search Based Apps in SharePoint Simple and Effective
February 20, 2013
New SharePoint users can often benefit from a hands on learning tutorial that lays out step by step how to get things done accurately and well, accurately. Let’s face it, in this fast paced world of ever evolving technology it is important to not waste time. Time is MONEY and money is the business.
Ontolica has done it again. With “Building Search Based Applications with SharePoint,” author Robert Piddocke walks users through an easy and effective way to create a search based application.
“Creating Search Based Applications in SharePoint is an easy and effective way to drive contextually valuable information to users without the limitations of having the documents in a specific library. Search can surface content from not just SharePoint but even from File Shares or other document management systems. All you need is the content crawled and you can create a document display mechanism based on search.”
The first step is to identify what metadata you want to use and if custom columns on SharePoint lists exist and can be used for your search. You must make a metadata mapping in the search service application for each piece of metadata you want to use. Run a full crawl.
Next up is creating a new page for your search based app. Then you must configure th core search result web parts but don’t forget that each web part needs a unique setting in order to function properly. After you adjust the layout xalt on the resut web parts to exclude the description and display relevant metadata you then have to set the sort on the result web part for date to display the most recent items and ply a query hat will match what you need users to see.
For more detailed information and some ridiculously helpful screenshots as well as a download we suggest heading over to Piddocke’s article.
Leslie Radcliff, February 20, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
-
- Subscribe to Beyond Search
Feature archive
News archive
-
