Identifying customers who are more likely to respond to new product offers is an important issue in direct marketing. In direct marketing, data mining has been used extensively to identify potential customers for a new product (target selection). Using historical purchase data, a predictive response model with data mining techniques was developed to predict a probability that a customer in Ebedi Microfinance bank will respond to a promotion or an offer. To achieve this purpose, a predictive response model using customers' historical purchase data was built with data mining techniques. The data were stored in a data warehouse to serve as management decision support system. The response model was built from customers' historic purchases and demographic dataset. Bayesian algorithm precisely Naïve Bayes algorithm was employed in constructing the classifier system. Both filter and wrapper feature selection techniques were employed in determining inputs to the model. The results obtained shows that Ebedi Microfinance bank can plan effective marketing of their products and services by obtaining a guiding report on the status of their customers which will go a long way in assisting management in saving significant amount of money that could have been spent on wasteful promotional campaigns.
Computational complexity is a characteristic of almost all Lesk-based algorithms for word sense disambiguation (WSD). In this paper, we address this issue by developing a simple and optimized variant of the algorithm using topic composition in documents based on the theory underlying topic models. The knowledge resource adopted is the English WordNet enriched with linguistic knowledge from Wikipedia and Semcor corpus. Besides the algorithm’s eficiency, we also evaluate its efectiveness using two datasets; a general domain dataset and domain-specific dataset. The algorithm achieves a superior performance on the general domain dataset and superior performance for knowledge-based techniques on the domain-specific dataset.
The majority of the state-of-the-art text categorization algorithms are supervised and therefore require prior training. Besides the rigor involved in developing training datasets and the requirement for repetition of training for different texts, working with multilingual texts poses additional unique challenges. One of these challenges is that the developer is required to have many different languages involved. Term expansion such as query expansion has been applied in numerous applications; however, a major drawback of most of these applications is that the actual meaning of terms is not usually taken into consideration. Considering the semantics of terms is necessary because of the polysemous nature of most natural language words. In this paper, as a specific contribution to the document index approach for text categorization, we present a joint multilingual/cross-lingual text categorization algorithm (JointMC) based on semantic term expansion of class topic terms through an optimized knowledge-based word sense disambiguation. The lexical knowledge in BabelNet is used for the word sense disambiguation and expansion of the topics' terms. The categorization algorithm computes the distributed semantic similarity between the expanded class topics and the text documents in the test corpus. We evaluate our categorization algorithm using a multilabel text categorization problem. The multilabel categorization task uses the JRC-Acquis dataset. The JRC-Acquis dataset is based on subject domain classification of the European Commission's EuroVoc microthesaurus. We compare the performance of the classifier with a model of it using the original class topics. Furthermore, we compare the performance of our classifier with two state-of-the-art supervised algorithms (each for multilingual and cross-lingual tasks) using the same dataset. Empirical results obtained on five experimental languages show that categorization with expanded topics shows a very wide performance margin when compared to usage of the original topics. Our algorithm outperforms the existing supervised technique, which used the same dataset. Cross-language categorization surprisingly shows similar performance and is marginally better for some of the languages.
Several language applications often require word semantics as a core part of their processing pipeline either as precise meaning inference or semantic similarity. Multi-sense embeddings (M-SE) can be exploited for this important requirement. M-SE seeks to represent each word by their distinct senses in order to resolve the conflation of meanings of words as used in different contexts. Previous works usually approach this task by training a model on a large corpus and often ignore the effect and usefulness of the semantic relations offered by lexical resources. However, even with large training data, coverage of all possible word senses is still an issue. In addition, a considerable percentage of contextual semantic knowledge are never learned because a huge amount of possible distributional semantic structures are never explored. In this paper, we leverage the rich semantic structures in WordNet using a graph-theoretic walk technique over word senses to enhance the quality of multisense embeddings. This algorithm composes enriched texts from the original texts. Furthermore, we derive new distributional semantic similarity measures for M-SE from prior ones. We adapt these measures to word sense disambiguation (WSD) aspect of our experiment. We report evaluation results on 11 benchmark datasets involving WSD and Word Similarity tasks and show that our method for enhancing distributional semantic structures improves embeddings quality on the baselines. Despite the small training data, it achieves state-of-the-art performance on some of the datasets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.