Semantic Retrieval of Trademarks Based on Conceptual Similarity

Anuar, Fatahiyah Mohd; Setchi, Rossitza; Lai, Yu‐Kun

doi:10.1109/tsmc.2015.2421878

Cited by 18 publications

(24 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Table 2 is the accuracy of the proposed system, which uses network architectures for testing. The definition of true positive rate (TPR) and the false positive rate (FPR) are written as formulas (8) and (9) Fig.7 shows the receiver operator characteristic (ROC) Curve of the overall and individual classes respectively. Area under the curve (AUC) is an indicator for evaluating system performance in ROC Curve.…”

Section: Resultsmentioning

confidence: 99%

“…Existing trademark search methods [4,5,6,7,8,9,10,11] have achieved high accuracy in similar trademark search work. However, news reports [15] [16] show that the confusion caused by optical illusion does affect the possibility of customers recognizing the wrong trademark and affecting the reputation and interests of trademark holders.…”

Section: Training and Testing Data Generation Pre-processingmentioning

confidence: 99%

“…In searching for similar shapes, work [7] regards high-frequency features as sketched images, and then uses the combination of grids as features to search for related images. Some works such as [8] and [9] used Tversky's similarity theory to search for the similarity between trademarks based on the semantic combination in trademarks. With the rise of deep learning today, some works utilized convolutional neural network (CNN) to recognize different types of trademarks from the past to the present [10], or classify trademarks by countries and region [11].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Trademark Background Analysis Method Based on Human Cognition of Visual Psychology

Hung

Chen

2020

Preprint

View full text Add to dashboard Cite

Trademark is a symbol that can be seen everywhere in life. In the case of effective use of trademark, people can quickly identify the organization, its image and reputation, so that the trademark owner can obtain resources for maintenance and development. However, there are unfair competitors in the market that would imitate similar trademarks to steal or damage the reputation and interests of the original trademark owners. Failure of preventing counterfeit trademarks from affecting the interests of others will make developers hard to develop novel technologies that contribute to the progress of the country and society due to insufficient or scarce resources. In the past research in this field,many studies have proposed protection methods for protecting trademark interests. These related studies have obtained excellent results in the search and analysis of trademarks. Nevertheless, some unfair competitors use the human cognition of visual psychology in the human vision system (HVS) to circumvent similar methods of trademark images to steal or damage the credit and interests of the original trademark owners. In order to deter unfair competitors or the abuse of such confusing trademarks, this article proposes a method for trademark background analysis. This article also proposes a method to generate a large amount of background trademark data. Then, we use that large amount of generated data for training neural networks that can analyze trademarks that are easily confused due to background. Eventually, we use the testing independent data for system verification. This experiment uses two common deep learning network architecture for testing. The experimental results show that the proposed system can achieve a true positive rate (TPR) over 98% in the face of this confusing trademark plagiarism method. And the comparison result is better than existing methods. This result confirms that the system proposed in this article can prevent infringement problems that used background.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Training and Testing Data Generation Pre-processingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Trademark Background Analysis Method Based on Human Cognition of Visual Psychology

Hung

Chen

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Moreover, some semantic deviations inevitably exist in a manual taxonomy such as WordNet, that is, not all locations of concepts in the "is-a" hierarchy of WordNet may always be the most appropriate ones compared with the cognitions of people, which may cause some deviations in the similarity measurements based on "is-a" relations. For example, the word pair food and fruit is given a low similarity of approximately 0.1 by existing algorithms based on "is-a" relations [2], [12]- [16], [18], whereas the human judgment yields a similarity of 0.77 (normalized) on the MC30 [48] dataset. Wikipedia is an online collaborative knowledge resource that has broad knowledge coverage and contains rich link semantics.…”

Section: Similarity Model Combining Wordnet and Wikipediamentioning

confidence: 99%

An Efficient Approach for Measuring Semantic Similarity Combining WordNet and Wikipedia

Liao

Zhang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The measurement of semantic similarity between concepts is an important research topic in natural language processing. In the past, several approaches for measuring the semantic similarity between concepts have been proposed based on WordNet or Wikipedia. However, improvements in the measurement accuracy of most methods have led to a dramatic increase in time complexity, and the existing methods do not effectively integrate WordNet and Wikipedia. In this paper, we focus on designing an efficient semantic similarity method based on WordNet and Wikipedia. To improve the accuracy of WordNet edgebased measures, we propose an edge weight model for combining edge and density information, which assigns a weight to each edge adaptively based on the number of direct hyponyms of the subsumer. Second, to improve the computational efficiencies of the existing Wikipedia link vector-based measures, we propose a new Wikipedia link feature-based semantic similarity method that converts Wikipedia links into semantic knowledge and replaces the TF-IDF statistical weight model in the existing measures. In addition, we propose two new word disambiguation strategies to further improve the accuracy of Wikipedia linkbased measures. Finally, to fully exploit the advantages of WordNet and Wikipedia, we propose two new aggregation schemas for combining WordNet "is-a" semantics and Wikipedia link semantics to replace the current aggregation schemas that combine WordNet "is-a" semantics with category semantics in Wikipedia. The experimental results show that our aggregation models are outstanding in terms of accuracy, efficiency and word coverage compared to state-of-the-art similarity measures.

show abstract

“…Similarity search based on DPMD algorithm. Similarity search has become increasingly important in large-scale and high-dimensional databases in which the contained objects do not possess any natural order [29]. The most common similarity search uses the mathematical notion of metric space, which enables efficient index structures to be built.…”

Section: 2mentioning

confidence: 99%

Anode effect prediction based on collaborative two-dimensional forecast model in aluminum electrolysis production

Chen¹,

Li²,

Chen³

et al. 2019

Journal of Industrial &Amp; Management Optimization

View full text Add to dashboard Cite

In this study, a new prediction algorithm is proposed, based on the collaborative two-dimensional forecast model (CTFM) that combines the traditional method and similarity search technique. The main idea of the algorithm is that the prediction of the change trend of the slope and the accumulated slope of the cell resistance as well as the useful knowledge obtained using the similarity search technique are used as the main criteria to calculate anode effect (AE)-prediction reliability. The accumulated mass deviation value is used as an auxiliary criterion to adjust the AE-prediction reliability. Finally, the current AE-process is marked according to the current AE-prediction reliability. The prediction model based on CTFM is tested on a real situation, in which multiple samples are extracted from the production of a 400 kA aluminum electrolysis cell. We observe that when the time advance of AE-prediction is within 20 ∼ 40 min, the accuracy rate of the CTFM algorithm is greater than 95% and the applicability of the method is excellent, showing a high prediction accuracy for different aluminum electrolysis cells.

show abstract

Semantic Retrieval of Trademarks Based on Conceptual Similarity

Cited by 18 publications

References 33 publications

Trademark Background Analysis Method Based on Human Cognition of Visual Psychology

Trademark Background Analysis Method Based on Human Cognition of Visual Psychology

An Efficient Approach for Measuring Semantic Similarity Combining WordNet and Wikipedia

Anode effect prediction based on collaborative two-dimensional forecast model in aluminum electrolysis production

Contact Info

Product

Resources

About