The LivingKnowledge Project
The LivingKnowledge Project is an effort by European researchers to develop a new kind of search technology, which takes into account factors such as context, bias, time, location and opinion.The team adopted a multi-disciplinary approach drawing on such diverse fields as computer science, social science, library science and semiotics. Diversity is an intrinsic part of knowledge. It is something that cannot be overlooked and something that must be evaluated when "searching" for resources.
The aim of this project is to track trends in public opinion about a topic, company or person as it has developed over time. The purpose of tracking trends is to automatically classify and organize opinions and biases in order to produce more insightful, better organized, and easier-to-understand output with high potential for social and economic development. The overall goal of the LivingKnowledge project is to bring a new quality into search and knowledge management technology, which makes search results more concise, complete and contextualised. The more ambitious goal is to be able to even predict the future.
This article appealed to me because I found that it covered topics we have been discussing for the past 3 weeks in class from the semantic web to categorization to contextualization to classification. In fact, the faceted classification system proposed by S. R. Ranganathan and documented in TDO Chapter 7 served as the inspiration for this project.
What is the problem and how does the LivingProject Solve it?
To begin with, the problem with existing internet search engines is that any search will yield results with little attention paid to categorization, context or bias. Incorporating these features could make "search" a better experience. In order to achieve this, the LivingKnowledge team turned to the faceted classification system and applied it to their database. As stated by Prof. Giunchiglia, the Team Co-Ordinator "we were able to turn Ranganathan's pseudo-algorithm into a computer algorithm and the computer scientists were able to use it to mine data from the web, extract its meaning and context, assign facets to it, and use these to structure the information based on the dimensions of diversity."
The next step was to spot biases and opinions of authors in different texts, another important piece of the puzzle. The team achieved this by analyzing and extracting meaning from both content and the way information is structured and laid-out, thereby adding another facet to the data.
The result of this effort is a new generation of search technology that supports the opinion-aware, diversity-aware and time-aware aggregation and exploration of knowledge.
Application Scenarios:
1. Media Content Analyser:
The LivingKnowledge system identifies social trends and monitors public opinion in both quantitative and qualitative terms. Analysing media content (both edited like traditional news and user generated like blogs) helps companies get a better insight about the public image of the company in relation to certain events such as the arrest of the CEO, or the effectiveness of a PR or advertising campaign, both current and past.
2. Future Predictor:
The team looked at “not only at current and past trends, but extrapolated them and drew on forecasts extracted from existing data to try to predict the future.” The future predictor application aims to provide functionality for "searching the future" by searching current news references, "predicting the future" by making inferences based on past occurrences and "mining the future" by searching for topics most relevant to a certain time segment.
It is clear that such predictive analysis, if it proves to be successful, could help change public policy for the betterment of all mankind, especially when predicting the future is going to depend on Big Data and how that data is interpreted and understood, not only by humans but more importantly by machines.
This indeed is a powerful tool in meeting that challenge.
http://www.telegraphindia.com/1120820/jsp/knowhow/story_15872307.jsp#.UHmYssXA_UF
- ashley.desouza's blog
- Login to post comments
