Welcome!

You will be redirected in 30 seconds or close now.

ColdFusion Authors: Yakov Fain, Jeremy Geelan, Maureen O'Gara, Nancy Y. Nee, Tad Anderson

Related Topics: ColdFusion

ColdFusion: Article

Data Classification Using a Digital Taxonomy

As the volume of digital content grows, so does the need for efficient and accurate data retrieval

As an organization's vast collection of data continues to grow, it becomes increasingly difficult for users to find the information they need. You need only to look at the success of Google to see the importance of search engine technology. Unfortunately, traditional search engines that rely primarily on keyword matching often return unintended results.

This makes finding the information that you're really after time consuming and inefficient. To make data search and discovery more productive, organizations are turning to taxonomy-based data classification.

Taxonomy classification is a means of creating order out of large collections of data. At its most basic level, taxonomy is simply a collection of terms or subjects. The strength of the model, however, comes from the taxonomy's ability to also define a term's relationship to other terms. This provides the means to derive the terms' context based on the relationships in the taxonomy. If a single term has several different meanings, it will have these additional associations defined in the taxonomy.

In most implementations the model is flexible, allowing for relationships to be expressed in much greater detail than is available from a strict hierarchical model. This allows for the definition of "related" and "equivalent" terms; something that is more difficult in typical hierarchical trees. Among other benefits, this makes it possible to implement an Amazon-like "recommendation" engine to find related items that are defined in similar topic areas within the same taxonomy.

When a taxonomy classification is being used, data that is added to a system is classified using the terms that have already been defined in the taxonomy. When data is associated with one or more terms, the data inherits the properties and relationships of those terms. This reduces the work involved with classifying new data. Also, as the taxonomy definition is improved and updated, the new data associations will be effective for the existing data without the need to go back and manually reclassify it.

Finding Your Data: No Problem!
Removing keyword ambiguity should be a goal of all search implementations. With conventional search engines, a keyword search for the term "star" could return results both on astronomy as well as Hollywood actors. With a taxonomy-based search, the multiple contexts would be known and presented to the user allowing for further refinement of search results based only on the desired subject matter, or "facets." Irrelevant data is filtered out, leaving behind only the results that are applicable to the selected topic area.

The same system that removes ambiguity also allows for the benefit of data discovery. Looking for data using a taxonomy navigation tool is similar to browsing the book aisles of a library. You may not know what you're looking for, but you'll know it when you see it. Users are able to "browse" the data that is associated with nearby terms in the taxonomy, allowing them to find information they might not have discovered in a search using known keywords.

Given that the same term could be relevant to many different subject areas, there are potentially many paths to the same data, allowing for expansive data discovery. Imagine searching for a brand of merlot red wine and then being presented a selection of foods that go best with that variety. That is the power of a taxonomy classification based search!

Enter ColdFusion and XML
XML is the emerging standard for defining taxonomies. Many of the currently available tools for creating a taxonomy specification provide XML export functionality. This is good news for ColdFusion developers, who already have a collection of functions available for working with XML.

In an XML definition, each term in the taxonomy is an element with its own collection of attributes and subelements. A standard definition will include tag markup for each type of relationship that can be represented. For most terms this will include "narrower term" and "broader term" tags, indicating the term's hierarchical position in a given context. In more advanced systems, XML elements would also be added to represent the nonhierarchical relationships. A sample XML specification is shown in Listing 1.

Although XML is widely used as the language for taxonomy definition, an authoritative-format standard for this definition is still pending. Given this, it is best to make the implementation as flexible as possible, allowing for future attributes and term relationship types to be added easily with little to no refactoring. Ideally, an accepted Document Type Definition (DTD) will be created that allows for the validation of the XML. Until then, it is possible to implement custom validation that uses XMLSearch() with an XPath expression to validate the required XML elements. Organizations may also want to consider creating their own DTD to be used for the validation.

Given a plain XML definition, it becomes trivial to use ColdFusion's XMLParse() to load the definition and create the XML object in memory. Once the XML object is obtained, an XPath expression can be used with ColdFusion's XMLSearch() to extract the relationships. Depending on the criteria specified in the XPath expression, it is possible to process a single specified term or the entire taxonomy at once. (See CFDJ, Vol. 4, issue 4 for an excellent article on parsing XML.)

Can ColdFusion Handle It?
Organizations that are most likely to implement a taxonomy classification system are those with high volumes of digital data. Given the large amount of data, it becomes important to keep performance in mind when designing the system. The two main factors that effect scalability are the number of terms in the taxonomy and the amount of data associated with those terms.

A taxonomy with 20,000 or more terms is generally considered large. Parsing the XML and storing the terms into memory are potentially intense processes. With ColdFusion, however, tests using a 100,000 term taxonomy on a mid-powered server resulted in load times of only a few seconds. This included the XMLParse() call to create the XML object, the XMLSearch() call to retrieve the terms, and multiple assignment calls to create an associative array of the terms along with the defined relationships. An additional step to perform the validation added only a marginal increase to the total processing time.

Even though a taxonomy will typically be limited to several thousand terms, there may be millions of data records associated with those terms. Once finalized, the size of the taxonomy definition tends to stay more or less fixed, unlike the data count of an active system, which will see continued growth. Even though reading the data associations from memory is fast, the memory consumption could become excessive as the system ages. It is usually safe to load the entire taxonomy into memory, because even a large classification will make only a modest dent in memory consumption. This is not true with the actual system data in which the typical design tradeoff between speed and memory must be considered.

To avoid scalability problems, you can rely on a simple CFQUERY database call to retrieve the associations given to a specific term. For further improvements, commonly referenced terms and their respective data associations can be cached for fast lookup. See Figure 1 for a basic process flow starting with the initial XML import, and concluding with the user obtaining results based on the specified criteria.

Future Development
The use of a taxonomy classification system for digital data is still relatively new. Over the past five years there has been much progress. However, work is still needed before there is a widely accepted vocabulary and common understanding of the framework and concepts.

One of the biggest challenges for an organization that wants to implement a taxonomy classification is the time and effort involved in creating the definition specification. Currently there are some commercially available definitions, but these are offered only in a limited number of business areas. Organizations that already have an institutional thesaurus are well positioned to use taxonomy-based classification. A thesaurus is often a precursor to a taxonomy, and the terms and vocabulary used to create a thesaurus are easily transferable. The National Information Standards Organization (NISO) has published guidelines for the construction of a Monolingual Thesauri, which is available in the ANSI Z39 specification. This is a good starting point for those exploring the possibility of implementing such a system.

Another implementation challenge is ensuring that data is classified correctly. There are auto-classification tools available that attempt to derive data context by using natural-language algorithms. These tools attempt to "understand" the content of the given data by evaluating not just the keywords, but also the circumstance. Once attained, the tools will assign the data to the proper term in the taxonomy. The accuracy of these tools won't match human classification but could be acceptable especially if the data is already tagged using some form of metadata.

Gaining Steam...
The idea of using a taxonomy to organize and classify data is not new. In fact, the term "taxonomy" comes from biology in reference to the classification of living things. Applying this idea to vast stores of digital content, however, is a practice that has only recently gained steam. As digital content repositories grow, finding your target data quickly and accurately will seem more like finding the proverbial needle in a haystack. A taxonomy classification system is an excellent complement to a traditional keyword search and will help users efficiently find the data they need.

More Stories By David Athey

David Athey is a senior developer with PaperThin Inc. based in Quincy, MA. He is an advanced Certified ColdFusion Developer with expertise in Enterprise Content Management and Web-based publishing.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


IoT & Smart Cities Stories
Apps and devices shouldn't stop working when there's limited or no network connectivity. Learn how to bring data stored in a cloud database to the edge of the network (and back again) whenever an Internet connection is available. In his session at 17th Cloud Expo, Ben Perlmutter, a Sales Engineer with IBM Cloudant, demonstrated techniques for replicating cloud databases with devices in order to build offline-first mobile or Internet of Things (IoT) apps that can provide a better, faster user e...
In his keynote at 19th Cloud Expo, Sheng Liang, co-founder and CEO of Rancher Labs, discussed the technological advances and new business opportunities created by the rapid adoption of containers. With the success of Amazon Web Services (AWS) and various open source technologies used to build private clouds, cloud computing has become an essential component of IT strategy. However, users continue to face challenges in implementing clouds, as older technologies evolve and newer ones like Docker c...
The Founder of NostaLab and a member of the Google Health Advisory Board, John is a unique combination of strategic thinker, marketer and entrepreneur. His career was built on the "science of advertising" combining strategy, creativity and marketing for industry-leading results. Combined with his ability to communicate complicated scientific concepts in a way that consumers and scientists alike can appreciate, John is a sought-after speaker for conferences on the forefront of healthcare science,...
Disruption, Innovation, Artificial Intelligence and Machine Learning, Leadership and Management hear these words all day every day... lofty goals but how do we make it real? Add to that, that simply put, people don't like change. But what if we could implement and utilize these enterprise tools in a fast and "Non-Disruptive" way, enabling us to glean insights about our business, identify and reduce exposure, risk and liability, and secure business continuity?
To Really Work for Enterprises, MultiCloud Adoption Requires Far Better and Inclusive Cloud Monitoring and Cost Management … But How? Overwhelmingly, even as enterprises have adopted cloud computing and are expanding to multi-cloud computing, IT leaders remain concerned about how to monitor, manage and control costs across hybrid and multi-cloud deployments. It’s clear that traditional IT monitoring and management approaches, designed after all for on-premises data centers, are falling short in ...
"The Striim platform is a full end-to-end streaming integration and analytics platform that is middleware that covers a lot of different use cases," explained Steve Wilkes, Founder and CTO at Striim, in this SYS-CON.tv interview at 20th Cloud Expo, held June 6-8, 2017, at the Javits Center in New York City, NY.
"MobiDev is a Ukraine-based software development company. We do mobile development, and we're specialists in that. But we do full stack software development for entrepreneurs, for emerging companies, and for enterprise ventures," explained Alan Winters, U.S. Head of Business Development at MobiDev, in this SYS-CON.tv interview at 20th Cloud Expo, held June 6-8, 2017, at the Javits Center in New York City, NY.
The deluge of IoT sensor data collected from connected devices and the powerful AI required to make that data actionable are giving rise to a hybrid ecosystem in which cloud, on-prem and edge processes become interweaved. Attendees will learn how emerging composable infrastructure solutions deliver the adaptive architecture needed to manage this new data reality. Machine learning algorithms can better anticipate data storms and automate resources to support surges, including fully scalable GPU-c...
As IoT continues to increase momentum, so does the associated risk. Secure Device Lifecycle Management (DLM) is ranked as one of the most important technology areas of IoT. Driving this trend is the realization that secure support for IoT devices provides companies the ability to deliver high-quality, reliable, secure offerings faster, create new revenue streams, and reduce support costs, all while building a competitive advantage in their markets. In this session, we will use customer use cases...
Machine learning has taken residence at our cities' cores and now we can finally have "smart cities." Cities are a collection of buildings made to provide the structure and safety necessary for people to function, create and survive. Buildings are a pool of ever-changing performance data from large automated systems such as heating and cooling to the people that live and work within them. Through machine learning, buildings can optimize performance, reduce costs, and improve occupant comfort by ...