This work proposes an efficient Summary Caption Technique that considers the multimodal summary and image captions as input to retrieve the correspondence images from the captions that are highly influential to the multimodal summary. Matching a multimodal summary with an appropriate image is a challenging task in computer vision and natural language processing. Merging in these fields is tedious, though the research community has steadily focused on cross-modal retrieval. These issues include the visual question-answering, matching queries with the images, and semantic relationship matching between two modalities for retrieving the corresponding image. Relevant works consider questions to match the relationship of visual information and object detection and to match the text with visual information and employing structural-level representation to align the images with the text. However, these techniques are primarily focused on retrieving the images to text or for image captioning. But less effort has been spent on retrieving relevant images for the multimodal summary. Hence, our proposed technique extracts and merge features in the Hybrid Image Text layer and captions in the semantic embeddings with word2vec where the contextual features and semantic relationships are compared and matched with each vector between the modalities, with cosine semantic similarity. In cross-modal retrieval, we achieve top five related images and align the relevant images to the multimodal summary that achieves the highest cosine score among the retrieved images. The model has been trained with seq-to-seq modal with 100 epochs, besides reducing the information loss by the sparse categorical cross entropy. Further, experimenting with the multimodal summarization with multimodal output dataset, in cross-modal retrieval, helps to evaluate the quality of image alignment with an image-precision metric that demonstrate the best results.

With the emergence of DNA microarray technology, scientists are continuously studying the expression levels of a large number of genes over different instances of time points. Analyzing the DNA microarray data confirm that the expression levels of two different genes vary simultaneously with the effect of external stimuli exhibiting different patterns. Therefore soft fuzzy clustering plays an important role in detecting patterns belonging to multiple clusters at the same time. Therefore this article proposes a novel fuzzy clustering technique utilizing weighted distance measure instead of the Euclidean distance using Grey Wolf Optimizer (GWO) as the global optimization techniques. Here clustering of microarray data is posed as a single objective optimization problem where the objective is to minimize the variability within the cluster and simultaneously maximizes the variability between the cluster. The newly proposed fuzzy-based weighted GWO clustering technique (Fuzzy-WDGWO) is then compared with some of the existing clustering techniques. Four different artificial datasets and three different real-life gene expression datasets have been considered to verify the efficiency of the proposed Fuzzy-WDGWO clustering technique both numerically as well as pictorially. Experimental analysis and performance evaluation of the proposed Fuzzy-WDGWO clustering technique show the superiority over the other existing clustering techniques such as PSO-FCM, FA-FCM, GWO-FCM, GWO-FCM, DE-FCM.

Microarray technology facilitates to monitor the expression levels of multiple genes simultaneously over a number of time points. For analyzing such microarray gene expression data, fuzzy clustering plays an important role. However, this fact has inspired us to propose a new automatic cluster detection method using modified differential evolution based fuzzy symmetric clustering technique. Here, allocation of gene data points to different clusters has been made based on a newly developed symmetry based cluster validity index. In order to get a best fuzzy partition of a gene dataset, FSym-index needs to be maximized. The search capability of the modified differential evolution based optimization technique in conjunction with the fuzzy symmetry based cluster validity index is applied to ascertain the good fuzzy partition as well as to obtain actual number of cluster present in a data set. The superiority of the proposed fuzzy-VMODEPS is compared with those obtained by some recent automatic clustering techniques, VGAPS, GCUK and HNGA over publicly available three real life gene expression data-sets.