Adjusted Cosine Similarity Research Articles

An integrated fuzzy adjusted cosine similarity and TOPSIS based recommendation system for information system requirements selection

Selecting the requirements for an information system based on their ranking values is one of the critical steps in software development. Various methods for computing the ranking order of the requirements have been developed using multi-criteria decision-making methods. This study shows selecting the requirements for different software releases using a recommendation system is an unresearched problem. We present a recommendation system (PRecSys) for computing the ranking order of the requirements in an information system using fuzzy-based adjusted cosine similarity measure and collaborative filtering to recommend the requirements to stakeholders with similar interests. The ranking order of the information systems requirements in PRecSys is computed using the fuzzy Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). The applicability of PRecSys is discussed by considering the requirements for a library information system. The PRecSys is compared with PHandler, an expert system for choosing the requirements, based on the following criteria: agreement measure, the number of judgments by decision-makers, and time complexity.

Fashion Recommendation System using Collaborative Filtering

Collaborative Filtering is an method used to build a recommendation system with the concept that conclusions from different clients are used to anticipate things that may be of interest to users. This research uses data from Rent the Runway and the method used is Item-based Collaborative filtering, where the system will look for similarities in products that have been purchased by customers and then look for predictive values. Fashion requires recommendations because it plays a crucial role in helping individuals express their identity, personal style, and personality through clothing choices, accessories, and dressing styles.The recommendation system uses the item method based on analyzing the number of purchases or sales and grouping according to each product category so that it can help consumers in choosing fashion products. It was found that the use of Adjusted Cosine Similarity produces better recommendations with an average MAE value of 0.2750, while Cosine Similarity with an average MAE difference of 0.3989. This proves that the use of adjusted cosine similarity can produce better recommendations because the adjustment algorithm not only considers user behavior, but also produces lower performance errors.

A similarity-based remaining useful life prediction method using multimodal degradation features and adjusted cosine similarity

When a large amount of full life-cycle data are available, similarity-based methods are the preferred method for remaining useful life (RUL) prediction due to their reliability and accuracy. Traditional similarity-based RUL prediction methods use a single model and single-scale degradation features, which are incapable of fully capturing the degradation behavior of the system. Additionally, the similarity of spatial orientation is neglected in the similarity-matching process. To fill these research gaps, a novel method is developed based on multimodal degradation features and adjusted cosine similarity (ACS) to tackle complex-system RUL prediction in this paper. Complete ensemble empirical mode decomposition with adaptive noise is employed to decouple global degradation and random fluctuations in run-to-failure sensor data. Slow feature analysis is utilized to obtain local degradation features, and residual terms are used as global degradation features. Then, multimodal degradation features are transformed into one-dimensional health degradation indicators by bidirectional gated recurrent unit autoencoder. An ACS is developed to estimate the matching similarity between the test degradation curve and the training degradation curve. The proposed scheme captures the time-varying multimodal degradation behavior and provides libraries of health curves with multiple degradation patterns. The designed scheme is evaluated on the C-MAPSS dataset and the results illustrate the competitiveness and effectiveness of the proposed method.

Modified Dynamic Movement Primitives: Robot Trajectory Planning and Force Control Under Curved Surface Constraints.

Dynamic movement primitives (DMPs) have been widely applied in robot motion planning and control. However, in some special cases, original discrete DMP fails to generalize proper trajectories. Moreover, it is difficult to produce trajectories on the curved surface. To solve the above problems, a modified DMP method is proposed for robot control by adding the scaling factor and force coupling term. First, the adjusted cosine similarity is defined to assess the similarity of the generalized trajectory with respect to the demonstrated trajectory. By optimizing the similarity, the trajectories can be generated in all situations. Next, by adding the force coupling term derived from adaptive admittance control to the transformation system of the original DMP, the controller achieves the force control ability. Then, the modified DMP-based robot control system is developed. The stability and convergence of the system are proved. Finally, the high precisions of the proposed method are verified by simulations and experiments. The method is significant for trajectory learning and generalization on the curved surface.

Detecting Urban Commercial Districts by Fusing Points of Interest and Population Heat Data with Region-Growing Algorithms

Reasonable urban commercial planning must clarify the location and scope of urban commercial districts (UCDs). However, existing studies typically detect spurious UCDs owing to the bias in a single data source while ignoring the continuity and ambiguity of commercial district boundaries. Therefore, in this study, we designed a two-stage approach for detecting UCDs. First, points of interest and population heat data were fused through hotspot and overlay analyses to detect core commercial areas. The boundaries of the UCDs were then identified by considering adjacent blocks using adjusted cosine similarity and region-growing algorithms. Finally, an experiment was conducted in Xiamen, revealing concentrated businesses on Xiamen Island and sparse businesses outside Xiamen Island. An experimental comparison with other strategies confirmed the improved modeling ability of this approach for the edge ambiguity of UCDs. This framework provides tools for urban commercial planning and helps recognize urban commercial patterns in a timely manner.

Color Polarization Image Super-Resolution Reconstruction via a Cross-Branch Supervised Learning Strategy

The division-of-focal-plane polarimeter, as a real-time polarization imaging technology, suffers from missing spatial resolution. To address this challenge, we propose a novel two-branch network for color polarization image super-resolution (CPSRNet), which provides a low-cost solution. Specifically, CPSRNet utilizes the degree of linear polarization (DoLP) image and intensity image (S0) as inputs instead of a conventional intensity image. The DoLP branch, as the high-frequency signal stream, supervises cross-branch feature activation in the shallow architecture and restores the polarization characteristic. CPSRNet, which mainly consists of a cross-branch activation module (CBAM) and a related-supervised feature residual fusion module (RSRFM), tends to improve the learning of high-frequency features both locally and globally. The proposed CBAM utilizes deeper features in the DoLP branch to activate the low-level high-frequency features in the intensity branch. Meanwhile, RSRFM introduces adjusted cosine similarity (ACOS) loss to fuse two-branch feature maps in a supervised-learning manner. ACOS loss exploits the intensity stream to encourage the fused feature stream to flow in the desired direction. In addition, we build a new dataset for the color polarization image super-resolution task. All the color polarization images are captured by a division-of-focal-plane polarization camera in real scenarios. We perform extensive experiments, including polarization image super-resolution and demosaicking, to confirm the superiority of our CPSRNet. The source code is available at https://github.com/yudadabing/CPSRNet.

An unsupervised learning damage diagnosis method based on virtual impulse response function and time series models

As it contains the inherent properties of the structure and is insensitive to variation of the input, the transmissibility function (TF) is widely used for structural damage detection under ambient excitation. However, in TF-based methods, the selection of the optimal damage sensitivity band often requires access to vibration data under different operating conditions, which is not conducive to unsupervised learning. In order to accommodate unsupervised learning and extract more damage information, this paper proposes a data-driven unsupervised damage detection method based on the virtual impulse response function (VIRF) and time series model (TSM). In this method, the VIRF is used as a damage-sensitive feature (DSF), the TSM is used for low-dimensional feature extraction of the VIRF, and the distance metric of the model coefficients is used to detect the occurrence of anomalies. The study analyses the type of TSM applied to VIRF and uses the parameter adjusted cosine similarity (PACS) as a more accurate statistical distance tool for the model coefficients. The application of two well-known benchmark structure experiments verifies the accuracy and practicality of the proposed method in damage detection under minor operational variations and ambient excitations. The comparative analysis and detection results indicate that the proposed method is advantageous in qualitatively quantifying the extent of the damage. Structural anomalies can be detected in both linear and nonlinear damage scenarios.

Analysing the impact of contextual segments on the overall rating in multi-criteria recommender systems

Depending on the RMSE and sites sharing travel details, enormous reviews have been posted day by day. In order to recognize potential target customers in a quick and effective manner, hotels are necessary to establish a customer recommender system. The data adopted in this study was rendered by the Trip Advisor which permits the customers to rate the hotel on the basis of six criteria such as, Service, Sleep Quality, Value, Location, Cleanliness and Room. This study suggest the multi-criteria recommender system to analyse the impact of contextual segments on the overall rating based on trip type and hotel classes. In this research we have introduced item-item collaborative filtering approach. Here, the adjusted cosine similarity measure is applied to identify the missing value for context in the dataset. For the selection of significant contexts the backward elimination with multi regression algorithm is introduced. The multi-collinearity among predictors is examined on the basis of Variance Inflation Factor (V.I.F). In the experimental scenario, the results are rendered based on hotel class and trip type. The performance of the multiregression model is evaluated by the statistical measures such as R-square, MAE, MSE and RMSE. Along with this, the ANOVA study is conducted for different hotel classes and trip types under 2, 3, 4 and 5 star hotel classes.

Classification of Stakeholders of an Information System using Fuzzy based Adjusted Cosine Similarity Measure during Requirements Elicitation Process

Requirements elicitation process is employed for the identification of stakeholders of an information system. A large number of stakeholders from different domains across the globe participate during the requirements elicitation process. Based on our review, we found that existing methods for the analysis of stakeholders do not support how to classify the stakeholders based on the similarity measures. These measures play an important role for the recommendation of an item/user in a recommendation system. Thus, the objective of this paper is to classify the stakeholders of an information system using fuzzy-based adjusted cosine similarity measure. We have identified the stakeholders of library information system and their opinions for different requirements are recorded. The identified stakeholders have been classified and analysed based on the following similarity measures under fuzzy environment, i.e., Cosine similarity measure, Euclidean distance, Pearson coefficient correlation etc., so that stakeholders having similar requirements can be recommended to strengthen the requirements elicitation process.

Adversarial Detection Based on Inner-Class Adjusted Cosine Similarity

Deep neural networks (DNNs) have attracted extensive attention because of their excellent performance in many areas; however, DNNs are vulnerable to adversarial examples. In this paper, we propose a similarity metric called inner-class adjusted cosine similarity (IACS) and apply it to detect adversarial examples. Motivated by the fast gradient sign method (FGSM), we propose to utilize an adjusted cosine similarity which takes both the feature angle and scale information into consideration and therefore is able to effectively discriminate subtle differences. Given the predicted label, the proposed IACS is measured between the features of the test sample and those of the normal samples with the same label. Unlike other detection methods, we can extend our method to extract disentangled features with different deep network models but are not limited to the target model (the adversarial attack model). Furthermore, the proposed method is able to detect adversarial examples crossing attacks, that is, a detector learned with one type of attack can effectively detect other types. Extensive experimental results show that the proposed IACS features can well distinguish adversarial examples and normal examples and achieve state-of-the-art performance.

Vehicle detection method for satellite videos based on enhanced vehicle features

Vehicle detection by satellite videos is important for urban intelligent traffic construction. A satellite videos vehicle detection method is proposed to solve the problems of the surrounding background and the vehicles having a similar color, and the low number of pixels occupied by vehicles. First, we design a weighted normalization sparse autoencoder network, highlighting the effective features and strengthening the expression of the vehicle features by introducing weighted normalization sparse autoencoders into the backbone. Second, we design a method for the early fusion of spatial information transmission, realizing the enhancement of spatial information in the deep feature map through atrous convolution. Finally, we propose K-means++ based on adjusted cosine similarity, stretching the anchor box size by introducing the adjusted cosine similarity calculation. The experimental results show that the average accuracy of this method in satellite videos vehicle detection reached 75.7%, and the recall rate reached 89.6%. This method can effectively detect vehicles in satellite videos, and the effect is significantly improved compared with other methods.

PENGGUNAAN METODE COLLABORATIVE FILTERING BASED UNTUK REKOMENDASI KENDARAAN BERMOTOR

Motorized vehicles are one of the main needs of every human being and also the most common form of transportation used by people. To choose a motorized vehicle, people should not choosing it in a hurry because it takes good consideration to choose the right brand, type, and the need for the vehicle. In making their choice, usually people will read the reviews from vehicle review sites such as Carmudi.co.id, OTO.com, KobaYogas.com, and so on. The purpose of this thesis is to help provide web-based vehicle recommendations using the values of rating and criteria selected by the user. User rating values are calculated with collaborative filtering. In addition to the rating value, users can also get vehicle recommendations by providing specifications of the vehicle needs. Rating values from the program users will be processed by using adjusted cosine similarity to determine their similarity score to the rating values from vehicle review sites and other users so the vehicle recommendations can be obtained according to the similarity of the other user ratings. Based on the results of User Acceptance Testing (UAT) from 21 respondents, the testing got an average score of 83.95% so the program can be categorized as “Very Good”.

RESTAURANT RECOMMENDER SYSTEM USING ITEM BASED COLLABORATIVE FILTERING AND ADJUSTED COSINE ALGORITHM SIMILARITY

In 2018, the Ministry of Industry (Kemenperin) stated that the food and beverage sector contributed 6.34% of the national gross domestic product (GDP). Currently, culinary information can be easily found, both in print and online. The amount of information available sometimes makes people over-informed, making it difficult to choose a restaurant based on their preferences. To assist consumers in selecting a restaurant, we need a system that can provide several recommendations. This study aims to implement the item-based Collaborative Filtering method using the Adjusted Cosine Similarity algorithm on a restaurant recommendation system. The test was carried out with 40 samples from UIN Syarif Hidayatullah Jakarta using purposive sampling because the sample was selected based on specific criteria, and 40 respondents can be said to be correct because of the minimum number of respondents is 30. The accuracy test uses precision, and the determination of the error value uses MAE. The analysis of the research results used three scenarios, which are 5, 20, and 40 users. The third scenario produces the best precision and MAE values. Precision is better if the precision value is close to 1, and MAE is getting better if the MAE value is getting closer to 0. So it can be concluded that the Item-Based method with the Adjusted Cosine algorithm has the best accuracy and error values when the number of users grows.

An optimized item-based collaborative filtering algorithm

Collaborative filtering over the years have emerged as an alternative recommender system to address some of the setbacks of content based filtering. Although, Collaborative filtering has offered some benefits to the majority of the online stores in recommending products to users using users’ ratings of similarity measure, its usage has also raised some doubt in the minds of researchers, regarding its effectiveness in handling ratings with limited number of users or no rating record from users. Thus, this has resulted in efforts by researchers in determining further ways to combat the issues attributed with the existing collaborative filtering techniques in terms of data sparsity or cold-start situations. This study focused on improving the traditional similarity measurements that currently exist on the item-based collaborative filtering, in order to accommodate and mitigate further the issue of cold-start situations. Thus, this study proposed an algorithm which is meant to balance the three current traditional measurement metrics such as: Cosine-based similarity, Pearson correlation similarity and Adjusted cosine similarity, in the direction of cold-start situations. The improved algorithm of traditional measurement metrics were further compared with the existing algorithm of the traditional metrics. Results showed that the proposed algorithm offered a better item-based collaborative filtering algorithm to recommendation systems than the existing, using data set from Movielens recommender system. Hence, the proposed algorithm did not only mitigate the drawbacks experienced with the three traditional algorithms in terms of data sparsity or cold-start situations but also retained the good features of the existing item-based collaborative filtering algorithm. Thus, the proposed algorithm complemented the strength of the three traditional measurement metrics with evidence shown when measured with Mean Absolute Error.

A New Semantic Similarity Measure Based on Ontology for Movie Rate Prediction

A recommendation algorithm comprises of two important steps: 1) Predicting rates, and 2) Recommendation. Rate prediction is a cumulative function of the similarity score between two movies and rate history of those movies by other users. There are various methods for rate prediction such as weighted sum method, regression, deviation based etc. All these methods rely on finding similar items to the items previously viewed/rated by target user, with assumption that user tends to have similar rating for similar items. Computing the similarities can be done using various similarity measures such as Euclidian Distance, Cosine Similarity, Adjusted Cosine Similarity, Pearson Correlation, Jaccard Similarity etc. All of these well-known approaches calculate similarity score between two movies using simple rating based data. Hence, such similarity measures could not accurately model rating behavior of user. In this paper, we will show that the accuracy in rate prediction can be enhanced by incorporating ontological domain knowledge in similarity computation. This paper introduces a new ontological semantic similarity measure between two movies. For experimental evaluation, the performance of proposed approach is compared with two existing approaches: 1) Adjusted Cosine Similarity (ACS), and 2) Weighted Slope One (WSO) algorithm, in terms of two performance measures: 1) Execution time and 2) Mean Absolute Error (MAE). The open-source Movielens (ml-1m) dataset is used for experimental evaluation. As our results show, the ontological semantic similarity measure enhances the performance of rate prediction as compared to the existing-well known approaches.

A collaborative filtering recommendation method based on discrete quantum-inspired shuffled frog leaping algorithms in social networks

In social network recommendation systems, the rating score prediction accuracy of the collaborative filtering (CF) method depends on both the extraction of the nearest neighbors and the calculation of user/project similarity. Based on a similar principle to user/project behavior, this paper uses the maximum intersection method to extract the optimal neighbor candidate set, and presents a weighted adjusted cosine similarity method to compute user/project similarity. Furthermore, to optimize the weights of the method, a novel optimization method called the discrete quantum-inspired shuffled frog leaping (DQSFL) algorithm is proposed, which is based on the shuffled frog leaping algorithm and quantum information theory. The DQSFL algorithm uses quantum movement equations to search for the optimal location according to the co-evolution of the quantum frog colony. The experiments demonstrate that the CF recommendation method based on DQSFL can effectively solve the rating data sparseness problem in the similarity computation process to improve the accuracy of the rating score prediction, and provide a better recommended result than traditional CF algorithms.

A cosine similarity based QUALIFLEX approach with hesitant fuzzy linguistic term sets for financial performance evaluation

The financial performance evaluation is important for investors of listed companies to make investment decisions. It enables investors to avoid haphazard investment and make a better decision in the capital market with high risk. Financial performance evaluation can be formulated as a kind of multi-criteria decision making (MCDM) problems with hesitant fuzzy linguistic term sets (HFLTSs). Considering that the number of criteria markedly exceeds the number of alternatives in real-world financial performance evaluation, this paper aims to develop a cosine similarity based QUALIFLEX (QUALItative FLEXible multiple criteria method) approach for MCDM with HFLTSs to effectively solve such problems. Firstly, a new fuzzy envelope of HFLTS is proposed using a Bonferroni mean (BM) operator. A new cosine similarity measure for HFLTSs is defined on the basis of trapezoidal membership functions of the new fuzzy envelopes. By taking a hesitancy index of HFLTS into account, an adjusted cosine similarity measure is developed for HFLTSs. Subsequently, a linear programming model is constructed to objectively determine the criteria weights on the basis of the adjusted cosine similarity measure. Then a cosine similarity based QUALIFLEX approach is put forward for MCDM with HFLTSs. Finally, an example of financial performance evaluation is analyzed to illustrate the effectiveness and applicability of the proposed approach.

EKF-GPR-Based Fingerprint Renovation for Subset-Based Indoor Localization with Adjusted Cosine Similarity.

Received Signal Strength Indicator (RSSI) localization using fingerprint has become a prevailing approach for indoor localization. However, the fingerprint-collecting work is repetitive and time-consuming. After the original fingerprint radio map is built, it is laborious to upgrade the radio map. In this paper, we describe a Fingerprint Renovation System (FRS) based on crowdsourcing, which avoids the use of manual labour to obtain the up-to-date fingerprint status. Extended Kalman Filter (EKF) and Gaussian Process Regression (GPR) in FRS are combined to calculate the current state based on the original fingerprinting radio map. In this system, a method of subset acquisition also makes an immediate impression to reduce the huge computation caused by too many reference points (RPs). Meanwhile, adjusted cosine similarity (ACS) is employed in the online phase to solve the issue of outliers produced by cosine similarity. Both experiments and analytical simulation in a real Wireless Fidelity (Wi-Fi) environment indicate the usefulness of our system to significant performance improvements. The results show that FRS improves the accuracy by 19.6% in the surveyed area compared to the radio map un-renovated. Moreover, the proposed subset algorithm can bring less computation.

PENERAPAN ALGORITMA COLABORATIVE FILTERING UNTUK PEMBERIAN REKOMENDASI PADA PRODUK KAIN TENUN (Studi Kasus Kain Tenun Parengan)

Electronic commerce (e-commerce) is a concept that can be described as a process of buying and selling goods on the internet or the process of buying and selling or exchange of products, services, and information through information networks including the Internet. In developing e-commerce should be able to provide recommendations to customers. This recommendation aims to provide an overview of information about products that are considered in accordance with the wishes of customers. The method used in this research is Item Based Collaborative Filtering, where the system will look for similarity purchase model (similarity item) With others. The system will search for ratings between items by level The similarities exist. Once the item-to-item rating is obtained, this rating will be used to calculate the value Similarity between items using the Adjusted Cosine Similarity approach. The last process is to calculate the rating prediction value that customers have never done to a particular item. This approach uses Weigted Sum formula results from predicted value will be made recommendations to customers

Online Performance Monitoring and Modeling Paradigm Based on Just-in-Time Learning and Extreme Learning Machine for a Non-Gaussian Chemical Process

A novel performance monitoring and online modeling method to deal with a non-Gaussian chemical process with multiple operating conditions is proposed. On the basis of the framework of the proposed method, a kernel extreme learning machine (ELM) technique is used to efficiently extract features from high dimensional process data. Additionally, the Fastfood kernel is introduced into kernel ELM to accelerate computation efficiency, which is relatively low at the prediction time. Then, a modified just-in-time learning (JITL) technique is applied for online modeling. In JITL, a novel similarity index, called modified adjusted cosine similarity (MACS), is proposed so as to improve the prediction performance of online modeling. The proposed paradigm provides an efficient, accurate, and fast approach to monitor and model the multimode chemical process. The validity and effectiveness are evaluated by applying the method to a synthetic non-Gaussian multimode model and the distillation system.