Search

Text mining is one of the branches of data mining and refers to as the computing process of finding new patterns and relations among datasets which appear not to be related. Data mining is an interdisciplinary field which uses statistics, artificial intelligence, and database systems to generate new tools for discovering patterns among datasets. Similarly, when dealing with textual data, we need to use various methods in different branches of computer science (e.g. linguistics) and statistics. This study reviews the techniques of text-based sentiment analysis pipeline including preprocessing, aspect extraction, feature selection, and classification techniques used by scholars recently. It also surveys different applications of semantic analysis in the context of social media, marketing, and product reviews.

Emotion is a primary aspect of communication and can be expressed in many modalities. Text-Based Emotion Detection (TBED), one of the fastest growing branches of Natural Language Processing (NLP), is the process of classifying syntactic or semantic units of a corpus into a given set of emotion classes proposed by a psychological model. Automatic TBED mechanisms use machine learning approaches to create computational platforms automating the process of extracting emotions. TBED has a wide variety of applications in the area of artificial intelligence: Semantic analysis of documents and public messages related to terrorist attacks (to mitigate risks), automated analysis of historical corpora, and study of product reviews (to assess customer satisfaction). This work reviews the current literature of TBED and the psychological models associated with them.

Real-Estate rent prediction in housing market analysis plays a key role in calculating the Rate of Return - a salient index used to evaluate real-estate investment options. Accurate rent prediction in real estate investment can help in generating capital gains and guaranty a financial success. In this paper, we carry out a comprehensive analysis and study of seven machine learning algorithms for rent prediction, including Linear Regression, Multilayer Perceptron, Random Forest, KNN, Locally Weighted Learning, SMO, and KStar algorithms. We train new model for the US territory, including three house types of single-family, townhouse, and condo. Each data instance in the dataset has 21 internal attributes (e.g., area space, price, number of bed/bathroom, rent, school rating, so forth). A subset of the collected features selected by filter methods for the prediction models. We also employ a hierarchical clustering approach to cluster the data based on two factors of house type, and average rent estimate of zip codes. The empirical results suggest that the rent prediction models based on lazy learning algorithms lead to higher accuracy and lower prediction error compared to eager learning methods.

Social media bots can change society's perspective in different aspects of life. This paper analyzes sentiment features and their effect on the accuracy of machine learning models for social media bot detection. Social bots can use tweet sentiment to create a backfire effect and confirmation bias to create a fake trend or change public opinion. We analyze bot detection problems based on sentiment features inspired by the work by Micheal Workman [1] and create new features based on textual information of online comments. We offer a quantitative approach to create new features and compare machine learning models for bot detection. This work is based on psychological and social effects inherent in tweets' text content based on the work by [1]. The new set of sentiment features are extracted from a tweet's text and used to train bot detection models. Also, we implement the new model for the Dutch language and achieve more than 87% accuracy for the Dutch tweets based on new sentiment features. Considering new sentiment features based on psychological and social factors for a tweet's text will open a potential research area for social media bot detection.

Ontology can describe real-world objects and relationships. It can be used for searching, web mining, and semantic analysis. However, creating the ontology for a different domain is a labor-intensive and computationally expensive task. This work introduces a new model to automate ontology creation for the real estate domain based on a natural language processing approach. The new model can save significant time and human labor in ontology creation in several domains. Also, the new model can enhance retrieval information systems in advanced semantic web search engines.

The real estate market creates one of the significant business domains for investors, but a wise investment in real estate is more important for low-income people who have just one home as their life-time investment. So during the economic recession, many homeowners lose their homes because of two major factors: one, they could not pay their mortgages, and second, the house could not be rented to cover the mortgages. So real estate investment can be secured if the house could be an excellent rental property. Information from real estate websites can be a rich source of knowledge essential to detect the house's potential as a rental property. This work uses a natural language processing approach to propose a new real estate investment model based on online textual information. For the first time, we apply a transfer learning approach based on multiple online resources to recognize the house as valuable rental property in the real estate market. Bidirectional attention models based on transformers (BERT) are used to extract features for semantic convolutional neural network models to secure real estate investment. This research has three main points: (1) using transformers to implement semantic CNN based on Airbnb, Zillow data (2) performance evaluation of traditional Machine learning models with our new transfer learning model for rent prediction. (3) a new public data set for more than 5 million houses in the U.S based on semantic information that can be used for real estate market research. This research offers a new model for safe investment in the real estate market based on the transfer learning approach.

Real estate investment decisions are critical for low-income people who have just one home as their life-time investment option. So during the COVID-19 pandemic, unemployment causes many homeowners with a low income to lose their homes because of two major factors: one, they could not pay their mortgages without a job, and second, their house could not be rented easily. Rent prediction in real-estate can guarantee the success of an investment. Online information from real estate websites plays a significant role in making a business decision to buy a home. This paper applies natural language processing models to introduce a new model for safe real estate investment based on online information. For the first time, we use a transfer learning model based on online information from various online resources to detect a profitable rental property. Bidirectional Encoder Representations from Transformers(BERT) are used to implement a semantic convolutional neural network model to predict real estate investment safety. This work introduces a new model for rent prediction based on the United States housing market. Our contribution is three-fold: (1) using natural language processing approach to use the semantics of online information on Airbnb, Zillow, Schools, Public transportation, and crime rate websites for rent prediction (2) We perform a comprehensive analysis of eager and lazy machine learning models as a traditional Machine learning models with our proposed new transfer learning model for rent prediction. (3) Creating a new public data set of semantic analysis for more than 5 million houses in the United States based on online information. This data set will be available for public research in natural language processing research for people analytic applications. This work introduces a new machine learning model to guarantee safe investment in the real estate market using a transfer learning approach based on online information.

Social Media provides an opportunity for people to share their idea about different aspects of life. Traveling is one of the essential aspects of life. In this paper, we use Bidirectional Encoder Representations from Transformers(BERT) for sentiment classification of online reviews to implement a Convolutional Neural Network model that can recommend airline tickets. Different sentiment classification models are compared based on people’s online reviews from six online social platforms. The new model can classify airline tickets as an economical choice or not economical choice ticket based on various aspects of the trips, including airline customer satisfaction, travel destination, hotel information, restaurants, and tourist attractions. This work contribution is two-fold: first, it examines the importance of choosing a transfer learning model for sentiment classification of online social platforms in the recommender system’s prediction accuracy. Second, the implementation of the Convolutional Neural Network model, which can classify airline tickets based on data generated by multiple online social platforms. Using the natural language processing approach to use the transfer learning model to improve the CNN model’s prediction accuracy is a new approach to use online social platforms to recommend economical airline tickets to consumers.

Millions of online posts about different topics and products are shared on popular social media platforms. One use of this content is to provide crowd-sourced information about a specific topic, event, or product. However, this use raises an important question: what percentage of the information available through these services is trustworthy? In particular, might some of this information be generated by a machine, i.e., a "bot" instead of a human? Bots can be, and often are, purposely designed to generate enough volume to skew an apparent trend or position on a topic, yet the consumer of such content cannot easily distinguish a bot post from a human post. This paper introduces a new model that uses Bidirectional Encoder Representations from Transformers (Google Bert) for sentiment classification of tweets to identify topic-independent features for the social media bot detection model. Using a Natural Language Processing approach to derive topic-independent features for the new bot detection model distinguishes this work from previous bot detection models. We achieve 94% accuracy classifying the contents of Cresci data set [1] as generated by a bot or a human, where the most accurate prior work achieved an accuracy of 92%.

Optimization of Quality of Service (QoS) of IEEE 802.11 Medium Access Control (MAC) protocol is a significant issue for Wireless Ad-hoc Networks; and during recent years many researches have been done to increase its performance and QoS. In this paper we focus on how to improve the throughput of EDCF mechanism and get the best QoS in frames transmission. We first outline the disadvantages of EDCF. Then we state an adjustment algorithm to resolve these drawbacks. The proposed adaptive scheme is used in the RTS/CTS-based method and adjusts Contention Window (CW) sizes based on network environment and finally enhances the network throughput.