Article Preview
TopIntroduction
In a series of papers(Atanassov, 1986)(Atanassov & Gargov, 1989) (K. T. Atanassov, 1994), Atanassov insinuate and developed the concept of Intuitionistic Fuzzy Sets(IFS) as the generalized version of the Fuzzy Sets(FS). Fuzzy Sets was proposed by Zadeh(Zadeh, 1965). IFSs are proficient in portraying the vague and imprecise data. Gau and Buehrer(Gau & Buehrer, 1993) in 1993 gave the notion of vague sets which was later proved to be similar as IFS by Bustine, H., and Burillo, P.(Burillo & Bustince, 1996) in 1996. Dengfeng and Chuntian(Dengfeng & Chuntian, 2002) proposed the similarity measures of IFSs and showed its applications to pattern recognitions but this measure have some counter-intuitive results which was shown by H.B. Mitchell(Mitchell, 2003). Also H.B. Mitchell(Mitchell, 2003) gave the modified version of similarity measures of Dengfeng and Chuntian(Dengfeng & Chuntian, 2002). Szmidt and Kacprzyk(Szmidt& Kacprzyk,2004)(Szmidt & Kacprzyk, 2005) proposed the similarity measure using the geometrical interpretation of IFSs(using the fact that inside triangle, any combination of the parameters characteristic foe elements belonging to an IFSs.) Hua-Wen Liu(H. W. Liu, 2005) invented the similarity measure overcoming the disadvantages of Li and Cheng(Dengfeng & Chuntian, 2002) and Hong and Kim(Hong & Kim, 1999). Li, Y., Olson, D. L., & Qin(Li et al.,2007) reviewed, analyzed the existing similarity measures and expressed their counter-intuitive examples. Hung and Yang(Hung & Yang, 2007) invented the similarity measure based on Lp metric and shown its applications in pattern recognition. On the basis of weighted Hausdorff distance, weighted Hamming distance and weighted Euclidean distance, some continuous distance and similarity measure are defined by Xu and Chen(Xu& Chen, 2008). Based on the cosine similarity measure for Fuzzy Sets Jun Ye(Ye, 2011) proposed the cosine similarity measure and weighted cosine similarity measure between IFSs and applied it to pattern recognition and medical diagnosis. Mo, Zhou and Song(Mo et al., 2012) gave the similarity measure which can be applied by government agencies to rescue in emergency scenario. A new similarity measure for IFSs induced by Sugeno integral was given by Hwang, Yang, Hung and Lee(Hwang et al., 2012). A new cotangent similarity measure between IFSs along with weighted fuzzy cotangent similarity and ordered weighted fuzzy cotangent similarity was proposed by Maoying(Maoying, 2013) and applied the results in medical diagnosis. Papakostas et al.(Papakostas et al., 2013) analyzed the existing similarity measure of IFSs in detail and applied them to artificial classification problems, medical diagnosis, pattern classification benchmarks and face recognition. Beliakov et al. (Beliakov et al., 2014) proposed a novel way for defining similarity measure for IFSs in which similarity measure is described by two components indicating similarity and hesitancy aspects and presented the applications of their method in Colorectal cancer diagnosis, Bacteria detection and image segmentation. Boran and Akay(Boran & Akay, 2014) provided the counter-intuitive cases of some previously proposed similarity measures and gave a new bi-parametric similarity measure with its applications in pattern recognition. Two different similarity measures were given by Peerasak Intarapaiboon(Intarapaiboon, 2014) in which first was set-theoretic measure and second combines the first measure and the notion of similarity measures on the notion of lattice. On the basis of transformation techniques, Chen and Chang(Chen & Chang, 2015) presented the similarity measures and applied the outcomes to pattern recognition. Deng, Jiang and Fu(Deng et al., 2015) investigated the monotonicity of the similarity measures of IFSs and defined three types of monotonic properties by analyzing the geometrical relation between IFSs. Nguyen(Nguyen, 2016) invented the knowledge based similarity measure and applied the outcomes to the pattern recognition. Many existing similarity measure failed to consider the abstention group influence. So, Zhou(Zhou, 2016) gave the new similarity measure by considering the abstention group influence and applied the outcomes in pattern recognition, multi-criteria group decision making and medical diagnosis. Deli and Cagman(Deli, 2016) invented four types of distances of IFSs and also gave similarity measure of IFSs. They built the decision making method based on similarity measure of IFSs and showed its applications in medical diagnosis. On the basis of Intuitionistic Fuzzy inclusion measure, a new similarity measure between IFSs was given by Beg and Rashid(Beg & Rashid, 2016) and applied it to multi-criteria decision making. Jun Ye(Ye, 2016) invented two new cosine similarity measure and weighted cosine similarity measure and applied it to on decision making problem by giving an example on choosing mechanical design schemes. On the grounds of double sequence and modulus function, Khan, Lohari, Mursaleen(Khan et al., 2017) gave a novel similarity measure and applied it to pattern recognition. Garg(Garg, 2018) pointed out the drawbacks of cosine similarity measure given by Ye and J.(Ye, 2011) by presenting the counter-intuitive examples and gave the improved version of cosine similarity measure with its applications in pattern recognition, medical diagnosis and multi-criteria group decision making. On the account of transformed isosceles triangle, a new similarity measure between IFSs was formed by Jiang, Jin, Lee and Yao(Jiang et al., 2019) and applied it to pattern recognition and clustering problems. The statistic ‘Jaccard index’ is found useful in comparing the similarity of sample sets and the concept of Jaccard index was used by Hwang, Yang and Hung(Hwang et al., 2018) to invent a new similarity measure. Ashraf, Khan, Lohani(Ashraf et al., 2019) invented a new similarity measure termed as Hybrid similarity measure which is the merger of Intuitionistic Fuzzy bounded variation(IFBV) and Intuitionistic Fuzzy metric based measures. Song, Huang, Wang and Quan(Song et al., 2019) proposed a similarity measure and illustrated it to medical diagnosis and cluster analysis. Singh and Kumar(Singh & Kumar, 2020) proposed a novel dice similarity measure for IFSs which is based on the inner product and implemented in pattern and face recognition. Recently in 2021, Harish Garg and Dimple Rani(Garg & Rani, 2021) presented a novel similarity measure which is formed from the concept of transformation of Right-angled techniques and its characteristics. To tackle with the machine learning problems, Saraswat and Gahlot(Saraswat & Gahlot, 2021) invented a new similarity measure and applied the results to pattern recognition.