Crying is the only way babies communicate about their communication or health status. Infant cry may be caused by medical or physical factors. Infant cry recognition and classification (ICRC) is the determination of various physical and physiological states of infants through computer-aided detection with the help of cry sound signals. This study is a review of the studies on the ICRC. To increase the scope of the study, the year limitation was not used in the literature review. While examining the studies in the literature, the datasets used, the feature sets, the classifier used, and the success rates obtained were examined in detail. After the review, the methods used for the ICRC were summarized to guide future studies, and an overview of the developments was presented to the researchers. In addition, the findings of the studies examined are given comparatively. As a result of this study, it has been determined that the most used traditional methods in ICRC are MFCC as a feature set and neural-network-based classifiers as classifiers. In recent years, it has been determined that mostly deep learning-based methods have started to be used.

Emotion recognition systems from speech signals are realized with the help of acoustic or spectral features. Acoustic analysis is the extraction of digital features from speech files using digital signal processing methods. Another method is the analysis of time-frequency images of speech using image processing. The size of the features obtained by acoustic analysis is in the thousands. Therefore, classification complexity increases and causes variation in classification accuracy. In feature selection, features unrelated to emotions are extracted from the feature space and are expected to contribute to the classifier performance. Traditional feature selection methods are mostly based on statistical analysis. Another feature selection method is the use of metaheuristic algorithms to detect and remove irrelevant features from the feature set. In this study, we compare the performance of metaheuristic feature selection algorithms for speech emotion recognition. For this purpose, a comparative analysis was performed on four different datasets, eight metaheuristics and three different classifiers. The results of the analysis show that the classification accuracy increases when the feature size is reduced. For all datasets, the highest accuracy was achieved with the support vector machine. The highest accuracy for the EMO-DB, EMOVA, eNTERFACE’05 and SAVEE datasets is 88.1%, 73.8%, 73.3% and 75.7%, respectively.

Classification is separating data into predefined categories by obtaining descriptive features. In the classification process, machine and deep learning algorithms assume that the class samples are evenly distributed. In particular, the dataset size used in deep learning is significant for classification success. However, obtaining balanced data distribution in real-life problems is very difficult. This negatively affects class-based accuracy. Various methods are used in the literature to overcome the unbalanced data problem. This study investigated the effects of GAN, SMOTE, and VAE methods on ECG data. For this purpose, the heartbeat signals in the MIT-BIH dataset were used. To test the performance of the methods, a performance comparison was made using real and synthetic data, and finally, the model trained with synthetic data was tested with real data. According to the results, 96.5% accuracy was obtained with the real data. The highest classification accuracy of 100.0% was obtained in VAE when using only synthetic data. In training with synthetic data and test results with real data, the highest classification success was 86.4% with SMOTE. When synthetic and real data sets are used together, the highest success rate is 98.6% with VAE. In addition, the accuracy of all classes is evenly distributed after data augmentation.

As in all fields, technological developments have started to be used in the field of medical diagnosis, and computer-aided diagnosis systems have started to assist physicians in their diagnosis. The success of computer-aided diagnosis methods depends on the method used; dataset, pre-processing, post-processing, etc. differ according to the processes. In this study, parameter optimization of support vector machines was performed with four different methods currently used in the literature to assist the physician in diagnosis. The success of each method was tested on two different Parkinson’s datasets and the results were compared within themselves and with the literature. According to the results obtained, the highest accuracy rates vary depending on the dataset and optimization method. While Improved Chaotic Particle Swarm Optimization achieved high success in the first dataset, Bat Algorithm achieved higher success in the other dataset. While the successful results obtained are better than some studies in the literature, they are at a level that can compete with some studies.

The evolution of conversational agents, in particular the case with chatbots, has experienced huge boosts in recent years, enabling a variety of tasks and allowing users to enjoy much more interaction. This research presents a sequential model for a Chatbot of multiple documents that is based on the best of the Llama2 mod-el. The document classification framework intends to offer a user-oriented as well as a versatile conversational approach that draws on data from several fields. Through proper implementation of state-of-the-art natural language processing technology, the chatbot can understand users' inquiries, retrieve the required in-formation from the uploaded files, and respond fluently and understandably. It provides document management processes, like file handling of PDF, DOCX, etc., which enables the user to work with almost all file types and formats. And that directly uses Hugging Face Transformers in such processes as text embed-ding and conversational generation. One of the key components of the system is the FAISS tool that allows for vector storage and retrieval keeping the chatbot operating efficiently in the process of searching and retrieving information from vast document collections. In summary, the work provided here lays out the foundations of the multi-doc system which is a powerful tool that can be used to improve the deployments and information search tasks, with the effect of boost-ing user engagement and productivity.

Liver is a vital organ of a body that has numerous functions. Severe malfunctioning of these functions causes a liver necrosis. The diagnosis of this necrosis is performed by a specialist who uses appropriate microscopic analysis. However, because this analysis is based on specialist's opinion, the results could be subjective and error prone. The aim of this study is to quantify the hepatic necrotic areas using computer aided algorithms and provide quantitative results that are not prone to errors.