Abstract: Atrademarks is a sign that you can use to distinguish your business goods or servicesfrom those of other traders. Trademark can be defined expressly in the form ofany symbol, logo, titles etc. so, they need to be secure. This projectdeciphers the hypothetic similarities among trademarks, which happens when morethan two or more trademarks hail equal or relevant semantic implant. Thestate-of-the-art by offering a semantic algorithm to similitude trademarks inpre conditions of hypothetic parallelism. By using data similarity, it isderived that search and indexing technique developed similarity distance. Theoffered reflow algorithm is confirmed using two resources: a trademark databaseof conflicting cases and a databases company names. Extends the conceptualmodel by developing and evaluating a semantic algorithm for trademark retrievalbased on conceptual similarity. The conceptual comparison of text documentsthat share similar domain, use similar concepts, or express similar ideas hasbeen studied extensively. The underlying technology embedded in existingtrademark search systems is primarily based on text-based retrieval. Use thedifferent domains to measures the accuracy of the algorithm which gathereddifferent data.
Keywords: Conceptual similarity, trademarkinfringement, trademark retrieval, trademark similarity.