Zum Inhalt
Zur Navigation

Trendontologies

Trendontologies improve the quality of automated trend detection. We used a knowledge-based approach and utilized formalized expert knowledge which we used to model three lightweight trend ontologies. We did this by the example of market research. This knowledge base is used to enrich the automatic trend detection process.

Trend

A trend, in terms of market research, is the evolution of customers' opinion referring to a specific topic that can be described by its categories or labels.Low level trendontology

Modelling Trend Ontology

For modelling the trend ontologies we relied on the experience of experts from the market research domain. We used their knowledge to build an initial set of keywords which were categorized by the main concepts of market research (we considered only the high tech market), which we modelled in RDFS using Protégé. The identified main categories are:
  • Image
  • Product quality
  • Customer relation
  • Service
  • Public opinion
  • Customers' view
  • Decision.
Each category is implemented as a class consisting of relevant concepts that describe the Medium level trendontologycategory. For example the ''product quality'' categorie has a set of concepts that contains ''reliability, performance, power'' etc. To each concept that has been classified by experts as being trend indicating we added the property ''trend-indicating''.
Because our keyword/concept based trend ontology is built on a very simple schema it can be easily applied to enrich for example the word based feature vector creation step of a machine learning method.

The semi-automatic analysis of relevant market research news done by experts resulted in three meta-level classes: ''general'', ''quantification'' and ''classification''. The ''general'' class includes groups of the most important concepts like ''supplier'' and ''company''. The ''quantification'' class contains the idea of identifiers and diversificators and adds the meta-concept ''amplifier''. ''classification'' consists of different categories that define the context of the quantifier. Its character is dynamic since it strongly depends on the context at a given point in time. An interesting subcategory of ''classification'' is the so called ''structure'' that defines the basic structure of the context. We observed that this category particularly refers to the economic model of the given market.
In the process of extending the keyword set we Meta level trendontologyobserved the emergence of so called ''semantic fields'' as of the ''Semantic Field Theory''.  Applying statistical methods (like term frequency in documents) combined with manual expertise we identified adjectives that are significant for the description of customers' opinion. The most relevant adjectives are:
  • reliable
  • competent
  • allround
  • up-to-date.
Conducting a search for the semantic fields of these adjectives led to the appearance of so called ''satisfier''s, ''disatisfier''s and ''sensitive''s. We then defined each main concept as a category with its semantic field and its identifier which consists of a ''diversificator''. Identificators are adjectives which belong to a concept and which describe its features. ''diversivicator'' defines ''satisfier'', ''disatisfier'' and ''sensitive'' which are adjectives grouped by their relevant meaning that refers to ''positive'', ''negative'' and ''neutral'' customers' opinion.

Similarity Matrix

Publications

Posters

Talks

Theses

  • Iavor Jelev: Preprocessing of documents for Emergent Trend Detection in text Collections, Diploma Thesis, FU Berlin
  • Mike Rohland: Generierung von semantischen Relationen aus Tags innerhalb der Folksonomien, Master Thesis, FU Berlin
  • Ievgeniia Ozeran: Trends und Web: Themen, Zeit und Texte, Master Thesis, FU Berlin
  • Diana Olivera: GUI für wissensbasierte Trendanalysen, Bachelor Thesis, FU Berlin
  • Lars Wißler: Trendontologien für wissensbasierte Trenderkennung - Erweiterung und Test, Bachelor Thesis, FU Berlin
  • Joachim Daiber: Candidate Selection and Evaluation in the Entity Extraction System DBPedia Spotlight, Bachelor Thesis, FU Berlin
  • Olga Streibel: Knowledge Based Trend Mining, Phd Thesis, FU Berlin

1st International Workshop on Mining the Future Internet!

MIFI2010 workshop

2012-12-15 10:33

Abschlussveranstaltung des Corporate Semantic Web Projekts

5 Jahre Corporate Semantic Web Abschlussveranstaltung am 16.1.2013

Weiterlesen …

2012-12-07 18:20

CSW active in OMG API4KB Standardization

API4KB is an initiative within OMG that aims at defining a standard programming interface for knowledge bases

Weiterlesen …

© 2008 FU Berlin | Feedback
This work has been partially supported by the  InnoProfile-Corporate Semantic Web project funded by the German Federal Ministry of Education and Research (BMBF) and the BMBF Innovation Initiative for the New German Länder - Entrepreneurial Regions.
doctor death jack kevorkianbuy metronidazole online no prescription