Archive for the ‘Knowledge Capture’ Category

ROCK, RACK And Knowledge Retention

Friday, April 18th, 2014

Roundtable on Knowledge Retention Techniques held on 21 May 2013.

From the post:

Back in May 2013, we held a Roundtable on Knowledge Retention Techniques. Carla Newman, Shaharudin Mohd Ishak and Ashad Ahmed very graciously shared with us their journey and experiences in Knowledge Retention.

Three videos, Carla Newman on ROCK (Retention of Critical Knowledge), Shaharudin Mohd Ishak on IE Singapore’s RACK (Retention of All Critical Knowledge), and Ashad Ahmend on Knowledge Retention.

Any knowledge problem of interest to Shell Oil Company is of interest to me! 😉

At what junctures in a knowledge retention process would topic maps have the greatest impact?

Not really interested in disrupting current approaches or processes but in discovering where topic maps could be a value add to existing systems.

Knowledge Leakage:..

Thursday, August 22nd, 2013

Knowledge Leakage: The Destructive Impact of Failing to Train on ERP Projects by Cushing Anderson.


This IDC study refines the concept of knowledge leakage and the factors that compound and mitigate the impact of knowledge leakage on an IT organization. It also suggests strategies for IT management to reduce the impact of knowledge leakage on organizational performance.

There is a silent killer in every IT organization — knowledge leakage. IT organizations are in a constant state of flux. The IT environment, the staff, and the organizational goals change continuously. At the same time, organizational performance must be as high as possible, but the impact of changing staff and skill leakage can cause 50% of an IT organization’s skills to be lost in six years.

“Knowledge leak is the degradation of skills over time, and it occurs in every organization, every time. It doesn’t discriminate based on operating system or platform, but it can kill organizational performance in as little as a couple of years.” — Cushing Anderson, vice president, IT Education and Certification research

I don’t have an IDC account so I can’t share with you what goodies may be inside this article.

I do think that “knowledge leakage” is a good synonym for “organizational memory.” Or should that be “organizational memory loss?”

I also don’t think that “knowledge leakage” is confined to IT organizations.

Ask the nearest supervisor that has had a long time administrative assistant retire. That’s real “knowledge leakage.”

The problem with capturing organizational knowledge, the unwritten rules of who to ask, for what and when, is that such rules are almost never written down.

And if they were, how would you find them?

Let me leave you with a hint:

The user writing down the unwritten rules needs to use their vocabulary and not one ordained by IT or your corporate office. And they need to walk you through it so you can add your vocabulary to it.

Or to summarize: Say it your way. Find it your way.

If you are interested, you know how to contact me.

Categorization of interestingness measures for knowledge extraction

Sunday, July 15th, 2012

Categorization of interestingness measures for knowledge extraction by Sylvie Guillaume, Dhouha Grissa, and Engelbert Mephu Nguifo.


Finding interesting association rules is an important and active research field in data mining. The algorithms of the Apriori family are based on two rule extraction measures, support and confidence. Although these two measures have the virtue of being algorithmically fast, they generate a prohibitive number of rules most of which are redundant and irrelevant. It is therefore necessary to use further measures which filter uninteresting rules. Many synthesis studies were then realized on the interestingness measures according to several points of view. Different reported studies have been carried out to identify “good” properties of rule extraction measures and these properties have been assessed on 61 measures. The purpose of this paper is twofold. First to extend the number of the measures and properties to be studied, in addition to the formalization of the properties proposed in the literature. Second, in the light of this formal study, to categorize the studied measures. This paper leads then to identify categories of measures in order to help the users to efficiently select an appropriate measure by choosing one or more measure(s) during the knowledge extraction process. The properties evaluation on the 61 measures has enabled us to identify 7 classes of measures, classes that we obtained using two different clustering techniques.

It will take some time to run down the original papers but I am curious in the mean time if:

  1. Anyone agrees or disagrees with the reduction of measures as having different names (page 10)?
  2. Anyone agrees or disagrees with the classification of measures into seven groups (pages 10-11)?

Knowledge Extraction and Consolidation from Social Media

Thursday, May 31st, 2012

Knowledge Extraction and Consolidation from Social Media KECSM2012 – November 11 – 12, Boston, USA.

Important dates

  • Jul 31, 2012: submission deadline full & short papers
  • Aug 21, 2012: notifications for research papers
  • Sep 10, 2012: camera-ready papers due
  • Oct 05, 2012: submission deadline poster & demo abstracts
  • Oct 10, 2012: notifications posters & demos

From the website:

The workshop aims to become a highly interactive research forum for exploring innovative approaches for extracting and correlating knowledge from degraded social media by exploiting the Web of Data. While the workshop’s general focus is on the creation of well-formed and well-interlinked structured data from highly unstructured Web content, its interdisciplinary scope will bring together researchers and practitioners from areas such as the semantic and social Web, text mining and NLP, multimedia analysis, data extraction and integration, and ontology and data mapping. The workshop will also look into innovative applications that exploit extracted knowledge in order to produce solutions to domain-specific needs.

We will welcome high-quality papers about current trends in the areas listed in the following, non-exhaustive list of topics. We will seek application-oriented, as well as more theoretical papers and position papers.

Knowledge detection and extraction (content perspective)

  • Knowledge extraction from text (NLP, text mining)
  • Dealing with scalability and performance issues with regard to large amounts of heterogeneous content
  • Multilinguality issues
  • Knowledge extraction from multimedia (image and video analysis)
  • Sentiment detection and opinion mining from text and audiovisual content
  • Detection and consideration of temporal and dynamics aspects
  • Dealing with degraded Web content

Knowledge enrichment, aggregation and correlation (data perspective)

  • Modelling of events and entities such as locations, organisations, topics, opinions
  • Representation of temporal and dynamics-related aspects
  • Data clustering and consolidation
  • Data enrichment based on linked data/semantic web
  • Using reference datasets to structure, cluster and correlate extracted knowledge
  • Evaluation of automatically extracted data

Exploitation of automatically extracted knowledge/data (application perspective)

  • Innovative applications which make use of automatically extracted data (e.g. for recommendation or personalisation of Web content)
  • Semantic search in annotated Web content
  • Entity-driven navigation of user-generated content
  • Novel navigation and visualisation of extracted knowledge/graphs and associated Web resources

I like the sound of “consolidation.” An unspoken or tacit goal of any knowledge gathering. Not much use in scattered pieces on the shop floor.

Collocated with the 11th International Semantic Web Conference (ISWC2012)

SIGKDD 2011 Conference

Tuesday, September 6th, 2011

A pair of posts from Ryan Rosario on the SIGKDD 2011 Conference.

Day 1 (Graph Mining and David Blei/Topic Models)

Tough sledding on Probabilistic Topic Models but definitely worth the effort to follow.

Days 2/3/4 Summary

Useful summaries and pointers to many additional resources.

If you attended SIGKDD 2011, do you have pointers to other reviews of the conference or other resources?

I added a category for SIGKDD.

Duolingo: The Next Chapter in Human Communication

Friday, April 29th, 2011

Duolingo: The Next Chapter in Human Communication

By one of the co-inventors of CAPTCHA and reCAPTCHA, Luis von Ahn, so his arguments should give us pause.

Luis wants to address the problem of translating the web into multiple languages.

Yes, you heard that right, translate the web into multiple languages.

Whatever you think now, watch the video and decide if you still feel the same way.

My question is how to adapt his techniques to subject identification?

Sixth International Conference on Knowledge Capture – K-Cap 2011

Monday, March 14th, 2011

Sixth International Conference on Knowledge Capture – K-Cap 2011

From the website:

In today’s knowledge-driven world, effective access to and use of information is a key enabler for progress. Modern technologies not only are themselves knowledge-intensive technologies, but also produce enormous amounts of new information that we must process and aggregate. These technologies require knowledge capture, which involve the extraction of useful knowledge from vast and diverse sources of information as well as its acquisition directly from users. Driven by the demands for knowledge-based applications and the unprecedented availability of information on the Web, the study of knowledge capture has a renewed importance.

Researchers that work in the area of knowledge capture traditionally belong to several distinct research communities, including knowledge engineering, machine learning, natural language processing, human-computer interaction, artificial intelligence, social networks and the Semantic Web. K-CAP 2011 will provide a forum that brings together members of disparate research communities that are interested in efficiently capturing knowledge from a variety of sources and in creating representations that can be useful for reasoning, analysis, and other forms of machine processing. We solicit high-quality research papers for publication and presentation at our conference. Our aim is to promote multidisciplinary research that could lead to a new generation of tools and methodologies for knowledge capture.


25 – 29 June 2011
Banff Conference Centre
Banff, Alberta, Canada

Call for papers has closed. Will try to post a note about the conference earlier next year.

Proceedings from previous conferences available through the ACM Digital Library – Knowledge Capture.

Let me know if you have trouble with the ACM link. I sometimes don’t get removal of all the tracing cruft off of URLs correct. There really should be a “clean” URL option for sites like the ACM.