Term Frequency-Inverse Document Frequency Model Research Articles

A Study on the Identification of the Water Army to Improve the Helpfulness of Online Product Reviews

Based on the perspective of identifying the water army, this paper uses the methods of machine learning and data visualization to analyze the helpfulness of online produce reviews, portray product portraits, and provide real and helpful product reviews. In order to identify and eliminate the water army, the Term Frequency-Inverse Document Frequency Model (TF-IDF) and Latent Semantic Index Model (LSI) are used. After eliminating the water army, three classification methods were selected to perform sentimental analysis, including logistics, SnowNLP, and Convolutional Neural Network for text(TextCNN). The TextCNN has the highest F1 score among the three classification methods. At the same time, the Latent Dirichlet Allocation mode (LDA) is used to extract the topics of various reviews. Finally, targeted countermeasures are proposed to manufacturers, consumers, and regulators.

Service priority classification using machine learning

This article details a procedure for classifying service cases with various priority levels based on machine learning (ML). It accurately defines the priority level of each service case. The presence of imbalanced datasets in service cases poses a challenge for achieving reliable classification accuracy. To address this, the use of the synthetic minority over-sampling technique (SMOTE) was proposed as the method for balancing the datasets prior to applying the ML method. From these experimental results, an improvement in the precision of the learning process was observed, which led to better outcomes in the test sets. This improvement was measured using the efficiency metrics from the confusion matrix. The experiment involved 6,182 service cases, categorized into four levels: critical, serious, moderate, and low. These were based on test comparisons with other ML methods. The accuracy achieved in the test data was 94.37%. By employing a hybrid technique to address the imbalance in SMOTE and the support vector machine model, it was found to be more effective than the comparative term frequency-inverse document frequency model that was used in conjunction with cosine similarity, which achieved an evaluation score of 70.14%.

Machine learning to predict notes for chart review in the oncology setting: a proof of concept strategy for improving clinician note-writing.

Leverage electronic health record (EHR) audit logs to develop a machine learning (ML) model that predicts which notes a clinician wants to review when seeing oncology patients. We trained logistic regression models using note metadata and a Term Frequency Inverse Document Frequency (TF-IDF) text representation. We evaluated performance with precision, recall, F1, AUC, and a clinical qualitative assessment. The metadata only model achieved an AUC 0.930 and the metadata and TF-IDF model an AUC 0.937. Qualitative assessment revealed a need for better text representation and to further customize predictions for the user. Our model effectively surfaces the top 10 notes a clinician wants to review when seeing an oncology patient. Further studies can characterize different types of clinician users and better tailor the task for different care settings. EHR audit logs can provide important relevance data for training ML models that assist with note-writing in the oncology setting.

Improving Helpdesk Chatbot Performance with Term Frequency-Inverse Document Frequency (TF-IDF) and Cosine Similarity Models

Helpdesk chatbots are growing in popularity due to their ability to provide help and answers to user questions quickly and effectively. Chatbot development poses several challenges, including enhancing accuracy in understanding user queries and providing relevant responses while improving problem-solving efficiency. In this research, we aim to enhance the accuracy and efficiency of the Helpdesk Chatbot by implementing the Term Frequency-Inverse Document Frequency (TF-IDF) model and the Cosine Similarity algorithm. The TF-IDF model is a method used to measure the frequency of words in a document and their occurrence in the entire document collection, while the Cosine Similarity algorithm is used to measure the similarity between two documents. After implementing and testing TF-IDF and Cosine Similarity models in the Helpdesk Chatbot, we achieved a 75% question recognition rate. To increase accuracy and precision, it is necessary to increase the knowledge dataset and improve pre-processing, especially in recognition and correct inaccurate spelling

Reliably Filter Drug-Induced Liver Injury Literature With Natural Language Processing and Conformal Prediction.

Drug-induced liver injury describes the adverse effects of drugs that damage the liver. Life-threatening results were also reported in severe cases. Therefore, liver toxicity is an important assessment for new drug candidates. These reports are documented in research papers that contain preliminary in vitro and in vivo experiments. Conventionally, data extraction from publications relies on resource-demanding manual labeling, which restricts the efficiency of the information extraction. The development of natural language processing techniques enables the automatic processing of biomedical texts. Herein, based on around 28,000 papers (titles and abstracts) provided by the Critical Assessment of Massive Data Analysis challenge, this study benchmarked model performances on filtering liver-damage-related literature. Among five text embedding techniques, the model using term frequency-inverse document frequency (TF-IDF) and logistic regression outperformed others with an accuracy of 0.957 on the validation set. Furthermore, an ensemble model with similar overall performances was developed with a logistic regression model on the predicted probability given by separate models with different vectorization techniques. The ensemble model achieved a high accuracy of 0.954 and an F1 score of 0.955 in the hold-out validation data in the challenge. Moreover, important words in positive/negative predictions were identified via model interpretation. The prediction reliability was quantified with conformal prediction, which provides users with a control over the prediction uncertainty. Overall, the ensemble model and TF-IDF model reached satisfactory classification results, which can be used by researchers to rapidly filter literature that describes events related to liver injury induced by medications.

Identifying single-item faked responses in personality tests: A new TF-IDF-based method.

Faking in a psychological test is often observed whenever an examinee may gain an advantage from it. Although techniques are available to identify a faker, they cannot identify the specific questions distorted by faking. This work evaluates the effectiveness of term frequency-inverse document frequency (TF-IDF)—an information retrieval mathematical tool used in search engines and language representations—in identifying single-item faked responses. We validated the technique on three datasets containing responses to the 10-item Big Five questionnaire (total of 694 participants, respectively 221, 243, and 230) in three faking situations. Each participant responded twice, once faking to achieve an objective in one of three contexts (one to obtain child custody and two to land a job) and once honestly. The proposed TF-IDF model has proven very effective in separating honest from dishonest responses—with the honest ones having low TF-IDF values and the dishonest ones having higher values—and in identifying which of the 10 responses to the questionnaire were distorted in the dishonest condition. We also provide examples of the technique in a single-case evaluation.

Scarcity-aware spam detection technique for big data ecosystem

To expand their business, companies in the industry use the big data ecosystem for handling enormous amounts of information. For this purpose, text data must be analyzed while ensuring data security and organizing authenticated and valuable data using spam filters. Several methods are available such as Word2Vec, bag-of-words, BERT, and term frequency-inverse document frequency (TF-IDF). However, none of these resolve the data scarcity issue that may result in the presence of incomplete information in collected documents. A technique that groups each document by subject and applies approximation using statistical methods is required to effectively solve this problem. This study proposes a natural language processing-based technique for spam detection that alters topics using a least-squares model and uses gradient-descent and altering-least-squares (AMALS) models to estimate missing data through TF-IDF and uniform-distribution. A performance evaluation demonstrates that the proposed technique outperforms 98% than the existing industrial TF-IDF model in predicting spam in big data ecosystems.

Keyword Extraction from Scientific Research Projects Based on SRP‐TF‐IDF

Keyword extraction by Term frequency-Inverse document frequency (TF-IDF) is used for text information retrieval and mining in many domains, such as news text, social contact text, and medical text. However, keyword extraction in special domains still needs to be improved and optimized, particularly in the scientific research field. The traditional TF-IDF algorithm considers only the word frequency in documents, but not the domain characteristics. Therefore, we propose the Scientific research project TF-IDF (SRP-TF-IDF) model, which combines TF-IDF with a weight balance algorithm designed to recalculate candidate keywords. We have implemented the SRP-TF-IDF model and verified that our method has better precision, recall, and F1 score than the traditional TF-IDF and TextRank methods. In addition, we investigated the parameter of our weight balance algorithm to find an optimal value for keyword extraction from scientific research projects.

SENTIMENT ANALYSIS OF CUSTOMER REVIEWS

Sentiment analysis is a process in which we classify text data as positive, negative, or neutral or into some other category, which helps understand the sentiment behind the data. Mainly machine learning and natural language processing methods are combined in this process. One can find customer sentiment in reviews, tweets, comments, etc. A company needs to evaluate the sentiment behind the reviews of its product. Customer sentiment can be a valuable asset to the company. This ultimately helps the company make better decisions regarding its product marketing and improving product quality. This paper focuses on the sentiment analysis of customer reviews from Amazon. The reviews contain textual feedback along with a rating system. The aim is to build a supervised machine learning model to classify the review as positive or negative. As reviews are in the text format, there is a need to vectorize the text to numerical format for the computer to process the data. To do this, we use the Bag-of-words model and the TF-IDF (Term Frequency-Inverse Document Frequency) model. These two models are related to each other, and the aim is to find which model performs better in our case. The problem in our case is a binary classification problem; the logistic regression algorithm is used. Finally, the performance of the model is calculated using a metric called the F1 score.

Impact of Different Approaches to Preparing Notes for Analysis With Natural Language Processing on the Performance of Prediction Models in Intensive Care.

Clinical note text was used to build machine learning models for adults admitted to the ICU. Preprocessing strategies studied were none (raw text), cleaning text, stemming, term frequency-inverse document frequency vectorization, and creation of n-grams. Model performance was assessed by the area under the receiver operating characteristic curve. Models were trained and internally validated on University of California San Francisco data using 10-fold cross validation. These models were then externally validated on Beth Israel Deaconess Medical Center data. ICUs at University of California San Francisco and Beth Israel Deaconess Medical Center. Ten thousand patients in the University of California San Francisco training and internal testing dataset and 27,058 patients in the external validation dataset, Beth Israel Deaconess Medical Center. None. Mortality rate at Beth Israel Deaconess Medical Center and University of California San Francisco was 10.9% and 7.4%, respectively. Data are presented as area under the receiver operating characteristic curve (95% CI) for models validated at University of California San Francisco and area under the receiver operating characteristic curve for models validated at Beth Israel Deaconess Medical Center. Models built and trained on University of California San Francisco data for the prediction of inhospital mortality improved from the raw note text model (AUROC, 0.84; CI, 0.80-0.89) to the term frequency-inverse document frequency model (AUROC, 0.89; CI, 0.85-0.94). When applying the models developed at University of California San Francisco to Beth Israel Deaconess Medical Center data, there was a similar increase in model performance from raw note text (area under the receiver operating characteristic curve at Beth Israel Deaconess Medical Center: 0.72) to the term frequency-inverse document frequency model (area under the receiver operating characteristic curve at Beth Israel Deaconess Medical Center: 0.83). Differences in preprocessing strategies for note text impacted model discrimination. Completing a preprocessing pathway including cleaning, stemming, and term frequency-inverse document frequency vectorization resulted in the preprocessing strategy with the greatest improvement in model performance. Further study is needed, with particular emphasis on how to manage author implicit bias present in note text, before natural language processing algorithms are implemented in the clinical setting.

Sustainable Fault Diagnosis of Imbalanced Text Mining for CTCS-3 Data Preprocessing

At present, the method for fault diagnosis and maintenance of the CTCS-3 (Chinese Train Control System Level 3) electronic equipment relies too heavily on expert knowledge. Moreover, the use of historical fault data is not valued. This paper proposes a sustainable fault diagnosis model based on imbalanced text mining. First, to process fault data from the field recorded in natural language, natural language processing technology is used to extract fault feature words. Then, a term frequency-inverse document frequency model is used to transform the fault feature words extracted from the database into vectors. It is worth noting that imbalance in the fault samples affects the accuracy of this sustainable fault diagnosis model. To solve this problem, we use the borderline-synthetic minority over-sampling technique in the step of predicting train fault components, we also use the backpropagation neural network we proposed and the naive Bayesian model which is commonly used as a classification model, to compare the prediction accuracy of these two algorithms. The experimental results perform well, which proves that the fault diagnosis method using the backpropagation neural network can further assist engineers to complete timely repair and maintenance work. The research in this paper has played a very important role in technical support for intelligent train dispatching and command, and will also play a positive role in technical support for the automatic operation of urban rail transit under the prevention and control of the new coronavirus.

An automatic software vulnerability classification framework using term frequency-inverse gravity moment and feature selection

Vulnerability classification is an important activity in software development and software quality maintenance. A typical vulnerability classification model usually involves a stage of term selection, in which the relevant terms are identified via feature selection. It also involves a stage of term-weighting, in which the document weights for the selected terms are computed, and a stage for classifier learning. Generally, the term frequency-inverse document frequency (TF-IDF) model is the most widely used term-weighting metric for vulnerability classification. However, several issues hinder the effectiveness of the TF-IDF model for document classification. To address this problem, we propose and evaluate a general framework for vulnerability severity classification using the term frequency-inverse gravity moment (TF-IGM). Specifically, we extensively compare the term frequency-inverse gravity moment, term frequency-inverse document frequency, and information gain feature selection using five machine learning algorithms on ten vulnerable software applications containing a total number of 27,248 security vulnerabilities. The experimental result shows that: (i) the TF-IGM model is a promising term weighting metric for vulnerability classification compared to the classical term-weighting metric, (ii) the effectiveness of feature selection on vulnerability classification varies significantly across the studied datasets and (iii) feature selection improves vulnerability classification.

COVID-19 Sensing: Negative Sentiment Analysis on Social Media in China via BERT Model.

Coronavirus disease 2019 (COVID-19) poses massive challenges for the world. Public sentiment analysis during the outbreak provides insightful information in making appropriate public health responses. On Sina Weibo, a popular Chinese social media, posts with negative sentiment are valuable in analyzing public concerns. 999,978 randomly selected COVID-19 related Weibo posts from 1 January 2020 to 18 February 2020 are analyzed. Specifically, the unsupervised BERT (Bidirectional Encoder Representations from Transformers) model is adopted to classify sentiment categories (positive, neutral, and negative) and TF-IDF (term frequency-inverse document frequency) model is used to summarize the topics of posts. Trend analysis and thematic analysis are conducted to identify characteristics of negative sentiment. In general, the fine-tuned BERT conducts sentiment classification with considerable accuracy. Besides, topics extracted by TF-IDF precisely convey characteristics of posts regarding COVID-19. As a result, we observed that people concern four aspects regarding COVID-19, the virus Origin (Gamey Food, 3.08%; Bat, 2.70%; Conspiracy Theory, 1.43%), Symptom (Fever, 2.13%; Cough, 1.19%), Production Activity (Go to Work, 1.94%; Resume Work, 1.12%; School New Semester Beginning, 1.06%) and Public Health Control (Temperature Taking, 1.39%; Coronavirus Cover-up, 1.26%; City Shutdown, 1.09%). Results from Weibo posts provide constructive instructions on public health responses, that transparent information sharing and scientific guidance might help alleviate public concerns.

Semantic-aware multi-keyword ranked search scheme over encrypted cloud data

Traditional searchable encryption schemes mostly adopt the TF-IDF (term frequency - inverse document frequency) model which ignores the semantic association between keywords and documents. It is a challenge to design an effective and secure semantic-aware search scheme. The topic model is based on “high-order co-occurrence”, i.e., how often words co-occur in different contexts. Hence, it can be utilized for modeling the latent semantics among texts. In this paper, we propose a novel privacy-persevering searchable encryption scheme based on the Latent Dirichlet Allocation (LDA) topic model. Documents are modeled by LDA, and the concept of topics is utilized to generate a document-topic relevance matrix and query topic vectors. The matrix is used as the index for the proposed scheme. The secure inner product operation is adopted to encrypt the index and query topic vectors, which provides accurate topic relevance score calculation between encrypted index and trapdoors. To improve the efficiency of our basic scheme, we adopt a special complete binary tree and use the “Greedy Depth First Search” algorithm. Our evaluation results demonstrate the effectiveness of our scheme.

Combination of Term Weighting and Integrated Color Intensity Co-occurrence Matrix for Two-Level Image Retrieval on Social Media Data

This paper proposes the two-level image retrieval that combines the text and image-based, to overcome the disadvantage of text or image as a query for image retrieval. In the text-based retrieval, three main steps are required. First, text and document pre-processing to retrieve words without affix, punctuation, and any stop words, to build the dictionary. Second, weighting the word from the dictionary, based on the frequency of words in text or document, using the Term Frequency-Inverse Document Frequency Model. Third, the similarity between a text query and the text document is calculated using Cosine Similarity. In the image-based approach for image retrieval, two main steps are required. First, feature extraction using Integrated Color Intensity Co-occurrence Matrix. This method will obtain two features at once, texture and color feature. Second, the similarity is calculated between an image query and database using Manhattan Distance. Social Media Data, Twitter, with Indonesian tweet and users, is used for the experiments. Image retrieval using a text, an image, and combination of both text and image, are compared in the experiments. The conducted experiments showed that the combination of text and image-based retrieval achieved the highest performance accuracy, compare with text or image-based retrieval.

Analysis of the characteristics of expressway traffic information propagation using Twitter

The purpose of this study was to analyze the characteristics of traffic information propagation via Twitter using a keyword analysis and a network analysis. For the keyword analysis, the main contents of Twitter messages were identified using a TF-IDF (Term Frequency - Inverse Document Frequency) model. For the network analysis, the network connectivity among Twitter users, including Traffic Information Producers (TIPs), Opinion Leaders (OLs), and their followers were measured by estimating the densities and mean distances in their follow networks. Based on the keyword analysis result, the words representing traffic conditions were revealed as the most influential keywords. In addition, the information regarding traffic accident occurrences was found to be most frequently retweeted. As a result of the network analysis, MBC news which is one of the biggest newscasts in Korea showed the greatest connectivity among TIPs. OLs proved more powerful in information propagation than TIPs. Conclusively, there is an apparent demand for establishing strategies to propagate traffic information based on the characteristics of Twitter in a more efficient manner.

A Heuristic Method for Learning Path Sequencing for Intelligent Tutoring System (ITS) in E-learning

Course sequencing is one of the vital aspects in an Intelligent Tutoring System (ITS) for e-learning to generate the dynamic and individual learning path for each learner. Many researchers used different methods like Genetic Algorithm, Artificial Neural Network, and TF-IDF (Term Frequency- Inverse Document Frequency) in E-leaning systems to find the adaptive course sequencing by obtaining the relation between the courseware. In this paper, heuristic semantic values are assigned to the keywords in the courseware based on the importance of the keyword. These values are used to find the relationship between courseware based on the different semantic values in them. The dynamic learning path sequencing is then generated. A comparison is made in two other important methods of course sequencing using TF-IDF and Vector Space Model (VSM) respectively, the method produces more or less same sequencing path in comparison to the two other methods. This method has been implemented using Eclipse IDE for java programming, MySQL as database, and Tomcat as web server.