Chatbot Tutorial 4 Utilizing Sentiment Analysis to Improve Chatbot Interactions by Ayşe Kübra Kuyucu Oct, 2024 DataDrivenInvestor

Analyzing multimodal data requires advanced techniques such as facial expression recognition, emotional tone detection, and understanding the impact between modalities. Sentiment analysis is a complex field and has played a pivotal role in the realm of data analytics. Ongoing advancements in sentiment analysis are designed for understanding and interpreting nuanced languages that are usually found in multiple languages, sarcasm, ironies, and modern communication found in multimedia data. Aspect-based analysis identifies the sentiment toward a specific aspect of a product, service, or topic. This technique categorizes data by aspect and determines the sentiment attributed to each. It is usually applied for analyzing customer feedback, targeting product improvement, and identifying the strengths and weaknesses of a product or service.

Bidirectional Encoder Representations from Transformers is abbreviated as BERT. It is intended to train bidirectional LSTM characterizations from textual data by conditioning on both the left and right context at the same time. As an outcome, BERT is fine-tuned just with one supplemental output layer to produce cutting-edge models for a variety of NLP tasks20,21.

• We will send each new chat message through TensorFlow’s pre-trained model to get an average Sentiment score of the entire chat conversation.
• By doing so, companies get to know their customers on a personal level and can better serve their needs.
• The number of social media users is fast growing since it is simple to use, create and share photographs and videos, even among people who are not good with technology.
• It collects and aggregates global word-to-word co-occurrences from the corpus for training, and it returns a linear substructure of all word vectors in a given space.

In English, words usually combine together to form other constituent units. Considering a sentence, “The brown fox is quick and he is jumping over the lazy dog”, it is made of a bunch of words and just looking at the words by themselves don’t tell us much. We use Sklearn’s classification_reportto obtain the precision, recall, f1 and accuracy scores. The DataLoader initializes a pretrained tokenizer and encodes the input sentences. We can get a single record from the DataLoader by using the __getitem__ function.

On a theoretical level, sentiment analysis innate subjectivity and context dependence pose considerable obstacles. Annotator bias and language ambiguity can all influence the sentiment labels assigned to YouTube comments, resulting in inconsistencies and uncertainties in the study. Python is a high-level programming language that supports dynamic semantics, object-oriented programming, and interpreter functionality. Deep learning approaches for sentiment analysis are being tested in the Jupyter Notebook editor using Python programming.

As a result, Table 1 depicts the labeled dataset distribution per proposed class. SpaCy stands out for its speed and efficiency in text processing, making it a top choice for large-scale NLP tasks. Its pre-trained models can perform various NLP tasks out of the box, including tokenization, part-of-speech tagging, and dependency parsing. Its ease of use and streamlined API make it a popular choice among developers and researchers working on NLP projects. I was able to repurpose the use of zero-shot classification models for sentiment analysis by supplying emotions as labels to classify anticipation, anger, disgust, fear, joy, and trust.

For instance, social media text is extremely nuanced and notoriously difficult for a machine learning algorithm to “understand”. ChatGPT is a GPT (Generative Pre-trained Transformer) machine learning (ML) tool that has surprised the world. Its breathtaking capabilities impress casual users, professionals, researchers, and even its own creators. Moreover, its capacity to be an ML model trained for general tasks and perform very well in domain-specific situations is impressive.

Adding sentiment analysis to natural language understanding, Deepgram brings in $47M

Finally, the above model is compiled using the ‘binary_crossentropy’ loss function, Adam optimizer, and accuracy metrics. NLP-based techniques have been used in standardized dialog-based systems such as Chat boxes11. Also, Text Analytics what is sentiment analysis in nlp is the most commonly used area where NLP is frequently used12. Machine learning algorithms with NLP can be used for further objectives like translating, summarizing, and extracting data, but with high computational costs.

From this, we obtained an accuracy of 94.74% using LSTM, 95.33% using BiLSTM, 90.76% using GRU, and 95.73% using the hybrid of CNN and BiLSTM. Generally, the results of this paper show that the hybrid of bidirectional RNN(BiLSTM) and CNN has achieved better accuracy than the corresponding simple RNN and bidirectional algorithms. As a result, using a bidirectional RNN with a CNN classifier is more appropriate and recommended for the classification of YouTube comments used in this paper.

Compared to XLM-T’s accuracy of 80.25% and mBERT’s 78.25%, these ensemble approaches demonstrably improve sentiment identification capabilities. The Google Translate ensemble model garners the highest overall accuracy (86.71%) and precision (80.91%), highlighting its potential for robust sentiment analysis tasks. The consistently lower specificity across all models underscores the shared challenge of accurately distinguishing neutral text from positive or negative sentiment, requiring further exploration and refinement. Compared to the other multilingual models, the proposed model’s performance gain may be due to the translation and cleaning of the sentences before the sentiment analysis task.

TextBlob’s sentiment analysis model is not as accurate as the models offered by BERT and spaCy, but it is much faster and easier to use. In this post, we will compare and contrast the four NLP libraries mentioned above in terms of their performance on sentiment analysis for app reviews. It supports multimedia content by integrating with Speech-to-Text and Vision APIs to analyze audio files and scanned documents. The tool can handle 242 languages, offering detailed sentiment analysis for 218 of them. Classify sentiment in messages and posts as positive, negative or neutral, track changes in sentiment over time and view the overall sentiment score on your dashboard. By highlighting these contributions, this study demonstrates the novel aspects of this research and its potential impact on sentiment analysis and language translation.

Accuracy of LSTM/GRU based architectures (created by Microsoft PowerPoint 2010). The old approach was to send out surveys, he says, and it would take days, or weeks, to collect and analyze the data. The very largest companies may be able to collect their own given enough time.

Development tools and techniques

The sentiment analysis system will note that the negative sentiment isn’t about the product but about the battery life. Finally, we applied three different text vectorization techniques, FastText, Word2vec, and GloVe, to the cleaned dataset obtained after finishing the preprocessing steps. The process of converting preprocessed textual data to a format that the machine can understand is called word representation or text vectorization. On October 7, Hamas launched a multipronged attack against Israel, targeting border villages and extending checkpoints around the Gaza Strip. The attack used armed rockets, expanded checkpoints, and helicopters to infiltrate towns and kidnap Israeli civilians, including children and the elderly1.

Figure 12a represents the graph of model accuracy when FastText plus LSTM model is applied. In the figure, the blue line represents training accuracy & the red line represents validation accuracy. Figure 12b represents the graph of model loss when FastText plus LSTM model is applied. In the figure, the blue line represents training loss & red line represents validation loss. The total positively predicted samples, which are already positive out of 27,727, are 18,097 & negative predicted samples are 5172. Similarly, true negative samples are 3485 & false negative samples are 973.

PyTorch is extremely fast in execution, and it can be operated on simplified processors or CPUs and GPUs. You can expand on the library with its powerful APIs, and it has a natural language toolkit. The biggest use case of sentiment analysis in industry today is in call centers, analyzing customer communications and call transcripts. That means that a company with a ChatGPT small set of domain-specific training data can start out with a commercial tool and adapt it for its own needs. There are also general-purpose analytics tools, he says, that have sentiment analysis, such as IBM Watson Discovery and Micro Focus IDOL. Therefore, LSTM, BiLSTM, GRU, and a hybrid of CNN and BiLSTM were built by tuning the parameters of the classifier.

The negative precision or the true negative accuracy reported 0.84 with the Bi-GRU-CNN architecture. In some cases identifying the negative category is more significant than the postrive category, especially when there is a need to tackle the issues that negatively affected the opinion writer. In such cases the candidate model is the model that efficiently discriminate negative entries. Another experiment was conducted to evaluate the ability of the applied models to capture language features from hybrid sources, domains, and dialects.

Hence, striking a record deal with the SEC means that Barclays and Credit Suisse had to pay a record value in fines. All of these issues imply a learning curve to properly use the (biased) API. Sometimes I had to do many trials until I reached the desired outcome with minimal consistency. In part 1 we represented each review as a binary vector (1s and 0s) with a slot/column for every unique word in our corpus, where 1 represents that a given word was in the review. So, simply considering 2-word sequences in addition to single words increased our accuracy by more than 1.6 percentage points. For our first iteration we did very basic text processing like removing punctuation and HTML tags and making everything lower-case.

Global NLP in Finance Market Size: Top-down Approach

Sentiment polarities of sentences and documents are calculated from the sentiment score of the constituent words/phrases. Most techniques use the sum of the polarities of words and/or phrases to estimate the polarity of a document or sentence24. The lexicon approach is named in the literature as an unsupervised approach because it does not require a pre-annotated dataset. It depends mainly on the mathematical manipulation of the polarity scores, which differs from the unsupervised machine learning methodology. The hybrid approaches (Semi-supervised or weakly supervised) combine both lexicon and machine learning approaches.

A hybrid parallel model that utlized three seprate channels was proposed in51. Character CNN, word CNN, and sentence Bi-LSTM-CNN channels were trained parallel. A positioning binary embedding scheme (PBES) was proposed to formulate contextualized embeddings that efficiently represent character, word, and sentence features. Binary and tertiary hybrid datasets were also used for the model assessment. The model performance was more evaluated using the IMDB movie review dataset.

How does NLP work?

However, for the experiment, this model was used in the baseline configuration and no fine tuning was done. Similarly, the dataset was also trained and tested using a multilingual BERT model called mBERT38. The experimental results are shown in Table 9 with the comparison of the proposed ensemble model. Hugging Face is a company that offers an open-source software library and a platform for building and sharing models for natural language processing (NLP).

The use of chatbots and virtual assistants powered by NLP is gaining popularity among financial institutions. These tools provide customers personalized financial advice and support, improving customer engagement and satisfaction. The total positively predicted samples which are already positive out of 20,795, are 13,081 & the negative predicted samples are 2,754. Similarly, true negative samples are 4,528 & false negative samples are 432. Figure 7b shows the plot of Loss between training samples & validation samples. Text Clustering and Topic Modelling are the two methods utilized most frequently to recognize topics included within a text corpus2.

And people usually tend to focus more on machine learning or statistical learning. But that often ends up in a lot of false positives, with a very obvious case being ‘good’ vs ‘not good’ — Negations, in general Valence Shifters. The data is not well balanced, and negative class has the least number of data entries with 6,485, and the neutral class has the most data with 19,466 entries. I want to rebalance the data so that I will have a balanced dataset at least for training.

Before we dive into the different methods for sentiment analysis, it’s important to note that it’s a technique within Natural Language Processing. Often called NLP, it is the study of how computers can understand human language. And although this is a specialty that is popular among Data Scientists, it’s not exclusive to the industry. In the secondary research process, various sources were referred for identifying and collecting information for this study. Secondary sources included annual reports, press releases, and investor presentations of companies; white papers, journals, and certified publications; and articles from recognized authors, directories, and databases.

You can foun additiona information about ai customer service and artificial intelligence and NLP. Hence, whether general domain ML models can be as capable as domain-specific models is still an open research question in NLP. A common next step in text preprocessing is to normalize the ChatGPT App words in your corpus by trying to convert all of the different forms of a given word into one. Stop words are the very common words like ‘if’, ‘but’, ‘we’, ‘he’, ‘she’, and ‘they’.

Literature review

From the figure, it is observed that training accuracy increases and loss decreases. So, the model performs well for offensive language identification compared to other pre-trained models. It’s a Stanford-developed unsupervised learning system for producing word embedding from a corpus’s global phrase co-occurrence matrix. The essential objective behind the GloVe embedding is to use statistics to derive the link between the words.

Do read the articles to get some more perspective into why the model selected one of them as the most negative and the other one as the most positive (no surprises here!). We can see that the spread of sentiment polarity is much higher in sports and world as compared to technology where a lot of the articles seem to be having a negative polarity. We, now, have a neatly formatted dataset of news articles and you can quickly check the total number of news articles with the following code. To find the class probabilities we take a softmax across the unnormalized scores. The class with the highest class probabilities is taken to be the predicted class. The id2label attribute which we stored in the model’s configuration earlier on can be used to map the class id (0-4) to the class labels (1 star, 2 stars..).

RNNs process chronological sequence in both input and output, or only one of them. According to the investigated problem, RNNs can be arranged in different topologies16. In addition to the homogenous arrangements composed of one type of deep learning networks, there are hybrid architectures combine different deep learning networks.

Sentiment analysis tools are essential to detect and understand customer feelings. Companies that use these tools to understand how customers feel can use it to improve CX. Sentiment analysis software notifies customer service agents — and software — when it detects words on an organization’s list. Sometimes, a rule-based system detects the words or phrases, and uses its rules to prioritize the customer message and prompt the agent to modify their response accordingly.

Taking this into account, we suggested using deep learning algorithms to find YouTube comments about the Palestine-Israel War, since the findings will help Palestine and Israel find a peaceful solution to their conflict. Section “Proposed model architecture” presents the proposed method and algorithm usage. Section “Conclusion and recommendation” concludes the paper and outlines future work. Organizations can enhance customer understanding through sentiment analysis, which categorizes emotions into anger, contempt, fear, happiness, sadness, and surprise8.

In the code above, we are building a functional React component to handle client side interaction with the Chat Application. Since we are using a functional component, we have access to React hooks, such as useState and useEffect. You can see the connection to the Socket server in useEffect, which will be called upon every re-render/on-load of the component. When a new message is emitted from the server, and event is triggered for the UI to receive and render that new message to all online user instances.

The proposed application proves that character representation can capture morphological and semantic features, and hence it can be employed for text representation in different Arabic language understanding and processing tasks. Meanwhile, many customers create and share content about their experience on review sites, social channels, blogs etc. The valuable information in the authors tweets, reviews, comments, posts, and form submissions stimulated the necessity of manipulating this massive data. The revealed information is an essential requirement to make informed business decisions. Understanding individuals sentiment is the basis of understanding, predicting, and directing their behaviours.

• One of the primary challenges encountered in foreign language sentiment analysis is accuracy in the translation process.
• Although, some researchers35 filter out the more numerous objective (neutral) phrases in the text and only evaluate and prioritise subjective assertions for better binary categorization.
• NLP powers AI tools through topic clustering and sentiment analysis, enabling marketers to extract brand insights from social listening, reviews, surveys and other customer data for strategic decision-making.
• BERT uses Transformers, and it learns the relation between a word to another word (or sub-words) in the given text of contextual nature.
• In this article, we will be working with text data from news articles on technology, sports and world news.

Besides, the detection of religious hate speech was analyzed as a classification task applying a GRU model and pre-trained word embedding50. The embedding was pre-trained on a Twitter corpus that contained different Arabic dialects. Supporting the GRU model with handcrafted features about time, content, and user boosted the recall measure. Deep learning applies a variety of architectures capable of learning features that are internally detected during the training process. The recurrence connection in RNNs supports the model to memorize dependency information included in the sequence as context information in natural language tasks14. And hence, RNNs can account for words order within the sentence enabling preserving the context15.

Social media platforms such as YouTube have sparked extensive debate and discussion about the recent war. As such, we believe that sentiment analysis of YouTube comments about the Israel-Hamas War can reveal important information about the general public’s perceptions and feelings about the conflict16. Moreover, social media’s explosive growth in the last decade has provided a vast amount of data for users to mine, providing insights into their thoughts and emotions17. Social media platforms provide valuable insights into public attitudes, particularly on war-related issues, aiding in conflict resolution efforts18.

Similarly, true negative samples are 7143 & false negative samples are 1222. The qualitative quality of the data and the enormous feedback volume are two obstacles in conducting customer feedback analysis. The analysis of textual comments, reviews, and unstructured text is far more complicated than the analysis of quantitative ratings, which can be done because ratings are quantitative. Nowadays, with the help of Natural Language Processing and Machine Learning, it is possible to process enormous amounts of text effectively without the assistance of humans.

For example if negative words are used in a review, the overall sentiment is not considered to be positive. With the spoken word, negative sentiment isn’t just about words, it’s also about tone. With the detectors the goal was to pull signals out of noise to help solve the mysteries of the universe.

In addition, LSTM models were widely applied for Arabic SA using word features and applying shallow structures composed of one or two layers15,40,41,42, as shown in Table 1. Another top option for sentiment analysis is VADER (Valence Aware Dictionary and sEntiment Reasoner), which is a rule/lexicon-based, open-source sentiment analyzer pre-built library within NLTK. A natural language processing (NLP) technique, sentiment analysis can be used to determine whether data is positive, negative, or neutral.

Learning generalizable AI models for multi-center histopathology image classification npj Precision Oncology

However, due to the massive scale of IR projects and the distribution of images, actual image datasets face an imbalance problem. As a result, the model still exhibits various overfitting phenomena during the training process. You can foun additiona information about ai customer service and artificial intelligence and NLP. Faced with massive image data, the huge computational workload and long training time still leave significant room for improvement in the timeliness of the model. Improvement ai based image recognition of recognition accuracy should also focus on the improvement of recognition efficiency, and should not satisfy the accuracy improvement and consume huge computational cost. In this regard, the study was carried out for the change optimization of the feature extraction module of DenseNet, and at the same time, the image processing adaptability of the parallel algorithm was improved.

Once the model’s outputs have been binarized, the underdiagnosis bias can be assessed by quantifying differences in sensitivity between patient races. Sensitivity is defined as the percentage of chest X-rays with findings that are identified as such by the AI model, whereas specificity is defined as the percentage of chest X-rays with no findings that are identified as such. The underdiagnosis bias identified by Seyyed-Kalantari et al. and reproduced here manifests in a higher sensitivity for white patients than for Asian and Black patients1. By substituting the amplitude of the source patch with that of the target patch.

Synthetic imagery sets new bar in AI training efficiency

Detection localizes and identifies the presence of organoids recognized by the model, providing the number of organoids that the model finds or misses compared to the ground truth. In the context of detection, OrgaExtractor detects organoids with a sensitivity of 0.838, a specificity of 0.769, and an accuracy of 0.813 (Fig. 2e). This research aims to introduce a unique Global Pooling Dilated CNN (GPDCNN) for plant disease identification (Zhang et al., 2019). Experimental evaluations on datasets including six common cucumber leaf diseases demonstrate the model’s efficacy.

In such areas, imaged based deep learning models for ECG recognition would serve best of which there are few studies in the literature. A recent paper created a model superior to signal based imaging achieving area under the received curve (AUROC) of 0.99 and area under Precision-Recall curve (AUPRC) 0.86 for 6 clinical disorders (Sangha et al., 2022). A machine learning-based automated approach (Suttapakti and Bunpeng, 2019) for classifying potato leaf diseases was introduced in a separate study. The maximum-minimum color difference technique was used alongside a set of distinctive color attributes and texture features to create this system. Image samples were segmented using k-means clustering and categorized using Euclidean distance.

We investigated several automated frameworks and models that have been proposed by researchers from across the world and are described in the literature. It is clear that AI holds great promise in the field of agriculture and, more specifically, in the area of plant disease identification. However, there is a need to recognize and solve the various issues that limit these models’ ability to identify diseases. In this part, we list the primary challenges that reduce the efficiency of automatic plant disease detection and classification. This research (Kianat et al., 2021) proposes a hybrid framework for disease classification in cucumbers, emphasizing data augmentation, feature extraction, fusion, and selection over three stages. The number of features is cut down with Probability Distribution-Based Entropy (PDbE) before a fusion step, and feature selection with Manhattan Distance-Controlled Entropy (MDcE) is done.

While the algorithm promises to excel in these types of sub-categorifications, Panasonic notes that this improved AI algorithm will help with subject identification and tracking in general when working in low light conditions. Frequent reversing operations of the Disconnecting Link often result in insufficient spring clamping force of the contact fingers and abrasion of the contact fingers. The local temperature maximum T1, T2, T3…Tn are obtained, the maximum value is selected as the hot spot temperature Tmax and the minimum value is selected as the normal temperature Tmin, and the relative temperature difference δt is obtained.

Incorporating the FFT-Enhancer in the networks boosts their performance

We specifically sought to develop strategies that were relatively easy to implement, could be adapted to other domains, and did not require knowledge of patient demographics during training or testing. The first approach consists of a data augmentation strategy based on varying the window width and field of view parameters during model training. This strategy aims to create a model that is robust to variations in these factors, for which the race prediction model exhibited patterns across different races.

The extraction of fiber feature information was more complete, and the IR effect has been improved6. To assist fishermen in managing the fishery industry, it needed to promptly eliminate diseased and dead fish, and prevent the transmission of viruses in fish ponds. Okawa et al. designed an abnormal fish IR model based on deep learning, which used fine-tuning to preprocess fish images appropriately. It was proved through simulation experiment that the abnormal fish IR model has improved the recognition accuracy compared to traditional recognition models, and the recall rate has increased by 12.5 percentage points7. To improve the recognition efficiency and accuracy of existing IR algorithms, Sun et al. introduced Complete Local Binary Patterns (CLBP) to design image feature descriptors for coal and rock IR.

What is AI? Everything to know about artificial intelligence

With the emergence of deep learning techniques, textile engineering has adopted deep networks for providing solutions to classification-related problems. These include classification based on fabric weaving patterns, yarn colors, fabric defects, etc.19,23. We investigated the performance of six deep learning architectures, which include VGG1624, VGG1924, ResNet5025, InceptionV326, InceptionResNetV227, and DenseNet20128. Each model is trained with annotated image repositories of handloom and powerloom “gamuchas”. Consequently, the features inherent to the fabric structures are ‘learned’, which helps to distinguish between unseen handloom and powerloom “gamucha” images.

Despite its advantages, the proposed method may face limitations in different tunnel construction environments. Varying geological conditions, diverse rock types, and environmental factors can affect its generalizability. Unusual mineral compositions or highly heterogeneous rock structures might challenge accurate image segmentation and classification. Additionally, input image quality, influenced by lighting, dust, or water presence, can impact performance.

Furthermore, we envision that an AI algorithm, after appropriate validation, could be utilized on diagnostic biopsy specimens, along with molecular subtype markers (p53, MMR, POLE). It is possible that with further refinement and validation of the algorithm, which can be run in minutes on the diagnostic slide image, that it could take the place of molecular subtype markers, saving time and money. First, the quality control framework, HistoQC81, generates a mask that comprises tissue regions exclusively and removes artifacts. Then, an AI model to identify tumor regions within histopathology slides is trained.

• The horizontal rectangular frame of the original RetinaNet has been altered to a rotating rectangular frame to accommodate the prediction of the tilt angle of the electrical equipment.
• It’s important to note that while the FFT-Enhancer can enhance images, it’s not always perfect, and there may be instances of noise artifacts in the output image.
• Summarizing all above, we can see that transfer learning has been shown to be an effective technique in improving the performance of computer vision models in various business applications.

Powdery mildew, downy mildew, healthy leaves, and combinations of these diseases were all included in the dataset. They used the cutting-edge EfficientNet-B4-Ranger architecture to create a classification model with a 97% success rate. Cucumbers, a much-loved and renewing vegetable, belong to the prestigious Cucurbitaceae family of plants.

Though we’re still a long way from creating Terminator-level AI technology, watching Boston Dyanmics’ hydraulic, humanoid robots use AI to navigate and respond to different terrains is impressive. GPT stands for Generative Pre-trained Transformer, and GPT-3 was the largest language model at its 2020 launch, with 175 billion parameters. The largest version, GPT-4, accessible through the free version of ChatGPT, ChatGPT Plus, and Microsoft Copilot, has one trillion parameters. The system can receive a positive reward if it gets a higher score and a negative reward for a low score.

• Second, we aimed to use the knowledge gained to reduce bias in AI diagnostic performance.
• Due to the dense connectivity, the DenseNet network enables feature reuse, which improves the algorithm’s feature representation and learning efficiency.
• The optimal time for capturing images is usually after blasting when the dust has settled and before the commencement of preliminary support work, as shown in Fig.

Our study introduces a novel deep learning model for automated loom type identification, filling a gap in existing literature and representing a pioneering effort in this domain. AI histopathologic imaging-based application within NSMP enables discernment of outcomes within the largest endometrial cancer molecular subtype. It can be easily added to clinical algorithms after performing ChatGPT App hysterectomy, identifying some patients (p53abn-like NSMP) as candidates for treatment analogous to what is given in p53abn tumors. Furthermore, the proposed AI model can be easier to implement in practice (for example, in a cloud-based environment where scanned routine H&E images could be uploaded to a platform for AI assessment), leading to a greater impact on patient management.

For the Ovarian, Pleural, and Bladder datasets, whole slide images (WSIs) serve as the input data. For computational tractability, we selected smaller regions ChatGPT from a WSI (referred to as patches) to train and build our model. More specifically, we extracted 150 patches per slide, with 1024 × 1024 pixels resolution.

The most common subtype (NSMP; No Specific Molecular Profile) is assigned after exclusion of the defining features of the other three molecular subtypes and includes patients with heterogeneous clinical outcomes. Shallow whole genome sequencing reveals a higher burden of copy number abnormalities in the ‘p53abn-like NSMP’ group compared to NSMP, suggesting that this group is biologically distinct compared to other NSMP ECs. Our work demonstrates the power of AI to detect prognostically different and otherwise unrecognizable subsets of EC where conventional and standard molecular or pathologic criteria fall short, refining image-based tumor classification.

The closer it is to red, the more likely it is to be classified as a ground truth label, while the closer it is to blue, the less likely it is. Heatmap analysis of samples (a, b) from the source domain and (c, d) from the target domain of the Ovarian cancer dataset. Despite its promising architecture, our evaluation of CTransPath’s impact on model performance yielded mixed outcomes. CTransPath achieved balanced accuracy scores of 49.41%, 69.13%, and 64.60% on the target domains of the Ovarian, Pleural, and Breast datasets, respectively, which were lower than the performance of AIDA on these datasets.

The model comprises different sized filters at the same layer, which helps obtain more exhaustive information related to variable-sized patterns. Moreover, Inception v3 is widely adopted in image classification tasks32,33 and is proved to achieve 78.1% accuracy with ImageNet Dataset and top-5 accuracy about 93.9%. The significance of this study lies in its potential to assist handloom experts in their identification process, addressing a critical need in the industry. By incorporating AI technologies, specifically deep learning models and transfer learning architectures, we aim to classify handloom “gamucha”s from power loom counterparts with cotton yarn type.