Abstract: In Japan, taxi is one of the popular transportations and taxi industry is one of the big businesses. However, in recent years, there has been a difficult problem of reducing the number of taxi drivers. In the taxi business, mainly three passenger catching methods are applied. One style is "cruising" that drivers catches passengers while driving on a road. Second is "waiting" that waits passengers near by the places with many requirements for taxies such as entrances of hospitals, train stations. The third one is "dispatching" that is allocated based on the contact from the taxi company. Above all, the cruising taxi drivers need the experience and intuition for finding passengers, and it is difficult to decide "the destination for cruising". The strong recommendation system for the cruising taxies supports the new drivers to find passengers, and it can be the solution for the decreasing the number of drivers in the taxi industry. In this research, we propose a method of recommending a destination for cruising taxi drivers. On the other hand, as a machine learning technique, the embedding models that embed the high dimensional data to a low dimensional space is widely used for the data analysis, in order to represent the relationship of the meaning between the data clearly. Taxi drivers have their favorite courses based on their experiences, and the courses are different for each driver. We assume that the course of cruising taxies has meaning such as the course for finding business man passengers (go around the business area of the city of go to main stations) and course for finding traveler passengers (go around the sightseeing places or big hotels), and extract the meaning of their destinations. We analyze the cruising history data of taxis based on the embedding model and propose the recommendation system for passengers. Finally, we demonstrate the recommendation of destinations for cruising taxi drivers based on the real-world data analysis using proposing method.
Abstract: Recently, online marketplaces in the e-commerce industry, such as Rakuten and Alibaba, have become some of the most popular online marketplaces in Asia. In these shopping websites, consumers can select purchase products from a large number of stores. Additionally, consumers of the e-commerce site have to register their name, age, gender, and other information in advance, to access their registered account. Therefore, establishing a method for analyzing consumer preferences from both the store and the product side is required. This study uses the Doc2Vec method, which has been studied in the field of natural language processing. Doc2Vec has been used in many cases to analyze the extraction of semantic relationships between documents (represented as consumers) and words (represented as products) in the field of document classification. This concept is applicable to represent the relationship between users and items; however, the problem is that one more factor (i.e., shops) needs to be considered in Doc2Vec. More precisely, a method for analyzing the relationship between consumers, stores, and products is required. The purpose of our study is to combine the analysis of the Doc2vec model for users and shops, and for users and items in the same feature space. This method enables the calculation of similar shops and items for each user. In this study, we derive the real data analysis accumulated in the online marketplace and demonstrate the efficiency of the proposal.
Abstract: With the increased utilization of technology devices around the world, healthcare and medical diagnosis are critical issues that people worry about these days. Doctors are doing their best to avoid any medical errors while diagnosing diseases and prescribing the wrong medication. Subsequently, artificial intelligence applications that can be installed on mobile devices such as rule-based expert systems facilitate the task of assisting doctors in several ways. Due to their many advantages, the usage of expert systems has increased recently in health sciences. This work presents a backward rule-based expert system that can be used for a headache diagnosis and medication recommendation system. The structure of the system consists of three main modules, namely the input unit, the processing unit, and the output unit.
Abstract: Cold-start is a notoriously difficult problem which
can occur in recommendation systems, and arises when there is
insufficient information to draw inferences for users or items. To
address this challenge, a contextual bandit algorithm – the Fast
Approximate Bayesian Contextual Cold Start Learning algorithm
(FAB-COST) – is proposed, which is designed to provide improved
accuracy compared to the traditionally used Laplace approximation
in the logistic contextual bandit, while controlling both algorithmic
complexity and computational cost. To this end, FAB-COST uses
a combination of two moment projection variational methods:
Expectation Propagation (EP), which performs well at the cold
start, but becomes slow as the amount of data increases; and
Assumed Density Filtering (ADF), which has slower growth of
computational cost with data size but requires more data to obtain an
acceptable level of accuracy. By switching from EP to ADF when
the dataset becomes large, it is able to exploit their complementary
strengths. The empirical justification for FAB-COST is presented, and
systematically compared to other approaches on simulated data. In a
benchmark against the Laplace approximation on real data consisting
of over 670, 000 impressions from autotrader.co.uk, FAB-COST
demonstrates at one point increase of over 16% in user clicks. On
the basis of these results, it is argued that FAB-COST is likely to
be an attractive approach to cold-start recommendation systems in a
variety of contexts.
Abstract: The world is expected to experience growth in the number of ageing population, and this will bring about high cost of providing care for these valuable citizens. In addition, many of these live with chronic diseases that come with old age. Providing adequate care in the face of rising costs and dwindling personnel can be challenging. However, advances in technologies and emergence of the Internet of Things are providing a way to address these challenges while improving care giving. This study proposes the integration of recommendation systems into homecare to provide real-time recommendations for effective management of people receiving care at home and those living with chronic diseases. Using the simplified Training Logic Concept, stakeholders and requirements were identified. Specific requirements were gathered from people living with cancer. The solution designed has two components namely home and community, to enhance recommendations sharing for effective care giving. The community component of the design was implemented with the development of a mobile app called Recommendations Sharing Community for Aged and Chronically Ill People (ReSCAP). This component has illustrated the possibility of real-time recommendations, improved recommendations sharing among care receivers and between a physician and care receivers. Full implementation will increase access to health data for better care decision making.
Abstract: Missing values in real-world datasets are a common
problem. Many algorithms were developed to deal with this
problem, most of them replace the missing values with a fixed
value that was computed based on the observed values. In
our work, we used a distance function based on Bhattacharyya
distance to measure the distance between objects with missing
values. Bhattacharyya distance, which measures the similarity of
two probability distributions. The proposed distance distinguishes
between known and unknown values. Where the distance between
two known values is the Mahalanobis distance. When, on the other
hand, one of them is missing the distance is computed based on the
distribution of the known values, for the coordinate that contains
the missing value. This method was integrated with Wikaya, a
digital health company developing a platform that helps to improve
prevention of chronic diseases such as diabetes and cancer. In order
for Wikaya’s recommendation system to work distance between users
need to be measured. Since there are missing values in the collected
data, there is a need to develop a distance function distances between
incomplete users profiles. To evaluate the accuracy of the proposed
distance function in reflecting the actual similarity between different
objects, when some of them contain missing values, we integrated it
within the framework of k nearest neighbors (kNN) classifier, since
its computation is based only on the similarity between objects. To
validate this, we ran the algorithm over diabetes and breast cancer
datasets, standard benchmark datasets from the UCI repository. Our
experiments show that kNN classifier using our proposed distance
function outperforms the kNN using other existing methods.
Abstract: Most movie recommendation systems have been developed for customers to find items of interest. This work introduces a predictive model usable by small and medium-sized enterprises (SMEs) who are in need of a data-based and analytical approach to stock proper movies for local audiences and retain more customers. We used classification models to extract features from thousands of customers’ demographic, behavioral and social information to predict their movie genre preference. In the implementation, a Gaussian kernel support vector machine (SVM) classification model and a logistic regression model were established to extract features from sample data and their test error-in-sample were compared. Comparison of error-out-sample was also made under different Vapnik–Chervonenkis (VC) dimensions in the machine learning algorithm to find and prevent overfitting. Gaussian kernel SVM prediction model can correctly predict movie genre preferences in 85% of positive cases. The accuracy of the algorithm increased to 93% with a smaller VC dimension and less overfitting. These findings advance our understanding of how to use machine learning approach to predict customers’ preferences with a small data set and design prediction tools for these enterprises.
Abstract: Rating prediction is an important problem for recommender systems. The task is to predict the rating for an item that a user would give. Most of the existing algorithms for the task ignore the effect of negative ratings rated by users on items, but the negative ratings have a significant impact on users’ purchasing decisions in practice. In this paper, we present a rating prediction algorithm based on factorization machines that consider the effect of negative ratings inspired by Loss Aversion theory. The aim of this paper is to develop a concave and a convex negative disgust function to evaluate the negative ratings respectively. Experiments are conducted on MovieLens dataset. The experimental results demonstrate the effectiveness of the proposed methods by comparing with other four the state-of-the-art approaches. The negative ratings showed much importance in the accuracy of ratings predictions.
Abstract: Cities are complex systems of diverse and inter-tangled activities. These activities and their complex interrelationships create diverse urban phenomena. And such urban phenomena have considerable influences on the lives of citizens. This research aimed to develop a method to reveal the causes and effects among diverse urban elements in order to enable better understanding of urban activities and, therefrom, to make better urban planning strategies. Specifically, this study was conducted to solve a data-recommendation problem found on a Korean public data homepage. First, a correlation analysis was conducted to find the correlations among random urban data. Then, based on the results of that correlation analysis, the weighted data network of each urban data was provided to people. It is expected that the weights of urban data thereby obtained will provide us with insights into cities and show us how diverse urban activities influence each other and induce feedback.
Abstract: With the gigantic increment in portable applications use and the spread of positioning and location-aware technologies that we are seeing today, new procedures and methodologies for location-based strategies are required. Location recommendation is one of the highly demanded location-aware applications uniquely with the wide accessibility of social network applications that are location-aware including Facebook check-ins, Foursquare, and others. In this paper, we aim to present a new methodology for location recommendation. The proposed approach coordinates customary spatial traits alongside other essential components including shortest distance, and user interests. We also present another idea namely, "activity trajectory" that represents trajectory that fulfills the set of activities that the user is intrigued to do. The approach dispatched acquaints the related distance value to select trajectory(ies) with minimum cost value (distance) and spatial-area to prune unneeded directions. The proposed calculation utilizes the idea of movement direction to prescribe most comparable N-trajectory(ies) that matches the client's required action design with least voyaging separation. To upgrade the execution of the proposed approach, parallel handling is applied through the employment of a MapReduce based approach. Experiments taking into account genuine information sets were built up and tested for assessing the proposed approach. The exhibited tests indicate how the proposed approach beets different strategies giving better precision and run time.
Abstract: Currently, there is excessively growing information
about places on Facebook, which is the largest social network but
such information is not explicitly organized and ranked. Therefore
users cannot exploit such data to recommend places conveniently and
quickly. This paper proposes a Facebook application and an Android
application that recommend places based on the number of check-ins
of those places, the distance of those places from the current location,
the number of people who like Facebook page of those places, and
the number of talking about of those places. Related Facebook data is
gathered via Facebook API requests. The experimental results of the
developed applications show that the applications can recommend
places and rank interesting places from the most to the least. We have
found that the average satisfied score of the proposed Facebook
application is 4.8 out of 5. The users’ satisfaction can increase by
adding the app features that support personalization in terms of
interests and preferences.
Abstract: Recommendation systems are widely used in
e-commerce applications. The engine of a current recommendation
system recommends items to a particular user based on user
preferences and previous high ratings. Various recommendation
schemes such as collaborative filtering and content-based approaches
are used to build a recommendation system. Most of current
recommendation systems were developed to fit a certain domain such
as books, articles, and movies. We propose1 a hybrid framework
recommendation system to be applied on two dimensional spaces
(User × Item) with a large number of Users and a small number
of Items. Moreover, our proposed framework makes use of both
favorite and non-favorite items of a particular user. The proposed
framework is built upon the integration of association rules mining
and the content-based approach. The results of experiments show
that our proposed framework can provide accurate recommendations
to users.
Abstract: The continued rise of e-commerce is the main driver of the rapid growth of global online purchase. Consumers can nearly buy everything they want at one occasion through online shopping. The purchase behavior models which focus on single product category are insufficient to describe online shopping behavior. Therefore, analysis of multi-category purchase gets more and more popular. For example, market basket analysis explores customers’ buying tendency of the association between product categories. The information derived from market basket analysis facilitates to make cross-selling strategies and product recommendation system.
To detect the association between different product categories, we use the market basket analysis with the multidimensional scaling technique to build an associated map which describes how likely multiple product categories are bought at the same time. Besides, we also build an inter-purchase time model for associated products to describe how likely a product will be bought after its associated product is bought. We classify inter-purchase time behaviors of multi-category products into nine types, and use a mixture regression model to integrate those behaviors under our assumptions of purchase sequences. Our sample data is from comScore which provides a panelist-label database that captures detailed browsing and buying behavior of internet users across the United States. Finding the inter-purchase time from books to movie is shorter than the inter-purchase time from movies to books. According to the model analysis and empirical results, this research finally proposes the applications and recommendations in the management.
Abstract: With a growing number of digital libraries and other
open education repositories being made available throughout the
world, effective search and retrieval tools are necessary to access the
desired materials that surpass the effectiveness of traditional, allinclusive
search engines. This paper discusses the design and use of
Folksemantic, a platform that integrates OpenCourseWare search,
Open Educational Resource recommendations, and social network
functionality into a single open source project. The paper describes
how the system was originally envisioned, its goals for users, and
data that provides insight into how it is actually being used. Data
sources include website click-through data, query logs, web server
log files and user account data. Based on a descriptive analysis of its
current use, modifications to the platform's design are recommended
to better address goals of the system, along with recommendations
for additional phases of research.
Abstract: A large number of semantic web service composition
approaches are developed by the research community and one is
more efficient than the other one depending on the particular
situation of use. So a close look at the requirements of ones particular
situation is necessary to find a suitable approach to use. In this paper,
we present a Technique Recommendation System (TRS) which using
a classification of state-of-art semantic web service composition
approaches, can provide the user of the system with the
recommendations regarding the use of service composition approach
based on some parameters regarding situation of use. TRS has
modular architecture and uses the production-rules for knowledge
representation.
Abstract: The proliferation of user-generated content (UGC) results in huge opportunities to explore event patterns. However, existing event recommendation systems primarily focus on advanced information technology users. Little work has been done to address novice and low-literacy users. The next billion users providing and consuming UGC are likely to include communities from developing countries who are ready to use affordable technologies for subsistence goals. Therefore, we propose a design framework for providing event recommendations to address the needs of such users. Grounded in information integration theory (IIT), our framework advocates that effective event recommendation is supported by systems capable of (1) reliable information gathering through structured user input, (2) accurate sense making through spatial-temporal analytics, and (3) intuitive information dissemination through interactive visualization techniques. A mobile pest management application is developed as an instantiation of the design framework. Our preliminary study suggests a set of design principles for novice and low-literacy users.
Abstract: Application of Expert System in the area of agriculture would take the form of Integrated Crop Management decision aids and would encompass water management, fertilizer management, crop protection systems and identification of implements. In order to remain competitive, the modern farmer often relies on agricultural specialists and advisors to provide information for decision-making. An expert system normally composed of a knowledge base (information, heuristics, etc.), inference engine (analyzes knowledge base), and end user interface (accepting inputs, generating outputs). Software named 'CROP-9-DSS' incorporating all modern features like, graphics, photos, video clippings etc. has been developed. This package will aid as a decision support system for identification of pest and diseases with control measures, fertilizer recommendation system, water management system and identification of farm implements for leading crops of Kerala (India) namely Coconut, Rice, Cashew, Pepper, Banana, four vegetables like Amaranthus, Bhindi, Brinjal and Cucurbits. 'CROP-9-DSS' will act as an expert system to agricultural officers, scientists in the field of agriculture and extension workers for decision-making and help them in suggesting suitable recommendations.
Abstract: In the past, there were more researches of recommendation system in applied electronic commerce. However, because all circles promote information technology integrative instruction actively, the quantity of instruction resources website is more and more increasing on the Internet. But there are less website including recommendation service, especially for teachers. This study established an instruction resource recommendation website that analyzed teaching style of teachers, then provided appropriate instruction resources for teachers immediately. We used the questionnaire survey to realize teacher-s suggestions and satisfactions with the instruction resource contents and recommendation results. The study shows: (1)The website used “Transactional Ability Inventory" that realized teacher-s style and provided appropriate instruction resources for teachers in a short time, it reduced the step of data filter. (2)According to the content satisfaction of questionnaire survey, four styles teachers were almost satisfied with the contents of the instruction resources that the website recommended, thus, the conception of developing instruction resources with different teaching style is accepted. (3) According to the recommendation satisfaction of questionnaire survey, four styles teachers were almost satisfied with the recommendation service of the website, thus, the recommendation strategy that provide different results for teachers in different teaching styles is accepted.
Abstract: The purpose of this paper is to explore the role of
cognitive decision effort in recommendation system, combined with
indicators "information quality" and "service quality" from IS success
model to exam the awareness of the user for the "recommended system
performance". A total of 411 internet user answered a questionnaire
assessing their attention of use and satisfaction of recommendation
system in internet book store. Quantitative result indicates following
research results. First, information quality of recommended system
has obvious influence in consumer shopping decision-making process,
and the attitude to use the system. Second, in the process of consumer's
shopping decision-making, the recommendation system has no
significant influence for consumers to pay lower cognitive
decision-making effort. Third, e-commerce platform provides
recommendations and information is necessary, but the quality of
information on user needs must be considered, or they will be other
competitors offer homogeneous services replaced.
Abstract: Recommender Systems act as personalized decision
guides, aiding users in decisions on matters related to personal taste.
Most previous research on Recommender Systems has focused on the
statistical accuracy of the algorithms driving the systems, with no
emphasis on the trustworthiness of the user. RS depends on
information provided by different users to gather its knowledge. We
believe, if a large group of users provide wrong information it will
not be possible for the RS to arrive in an accurate conclusion. The
system described in this paper introduce the concept of Testing the
knowledge of user to filter out these “bad users".
This paper emphasizes on the mechanism used to provide robust
and effective recommendation.