Empirical comparison of supervised learning techniques for missing value imputation

Knowledge and Information Systems - Many data mining algorithms cannot handle incomplete datasets where some data samples are missing attribute values. To solve this problem, missing value...     

Assessing the helpfulness of online hotel reviews: A classification-based approach

With the rapid development of Web 2.0, travelers have started sharing their travel experiences on websites. The expanding amount of online hotel reviews results in the problem of information overload. Therefore, the effective identification of helpful reviews has become an important research issue. In this study, online hotel reviews were collected from TripAdvisor.com, and the helpfulness of these reviews was comprehensively investigated from the aspects of review quality, review sentiment, and reviewer characteristics. Review helpfulness prediction models were also developed by using classification techniques. The results indicate that reviewer characteristics are good predictors of review helpfulness, whereas review quality and review sentiment are poor predictors of review helpfulness.     

Dynamic resource allocation with beamforming for MIMO OFDM systems: performance and effects of imperfect CSI

We propose three different dynamic resource allocation algorithms using adaptive beamforming for multiple-input multiple-output (MIMO) OFDM systems, and investigate their performance over multipath fading channels under perfect and imperfect channel state information (CSI). These approaches involve the use of adaptive modulation, adaptive frequency-domain power allocation, and/or adaptive sub-channel allocation. By employing the proposed approaches in MIMO/OFDM systems, significant performance improvement can be achieved compared to the conventional adaptive antenna array based OFDM. The investigation of the effects of imperfect CSI reveals that the adaptive-modulation based approach is too sensitive to channel estimation errors, and that its performance is worse than the adaptive frequency-domain power allocation and/or adaptive sub-channel allocation approaches. The performance analysis also shows that combining adaptive power allocation with sub-channel allocation yields the best performance under imperfect CSI while being robust to channel estimation errors.     

Bovine Induced Pluripotent Stem Cells Are More Resistant to Apoptosis than Testicular Cells in Response to Mono-(2-ethylhexyl) Phthalate

Although the androgen receptor (AR) has been implicated in the promotion of apoptosis in testicular cells (TSCs), the molecular pathway underlying AR-mediated apoptosis and its sensitivity to environmental hormones in TSCs and induced pluripotent stem cells (iPSCs) remain unclear. We generated the iPSCs from bovine TSCs via the electroporation of OCT4. The established iPSCs were supplemented with leukemia inhibitory factor and bone morphogenetic protein 4 to maintain and stabilize the expression of stemness genes and their pluripotency. Apoptosis signaling was assessed after exposure to mono-(2-ethylhexyl) phthalate (MEHP), the active metabolite of di-(2-ethylhexyl) phthalate. Here, we report that iPSCs were more resistant to MEHP-induced apoptosis than were original TSCs. MEHP also repressed the expression of AR and inactivated WNT signaling, and then led to the commitment of cells to apoptosis via the cyclin dependent kinase inhibitor p21^CIP1. The loss of the frizzed receptor 7 and the gain of p21^CIP were responsible for the stimulatory effect of MEHP on AR-mediated apoptosis. Our results suggest that testicular iPSCs can be used to study the signaling pathways involved in the response to environmental disruptors, and to assess the toxicity of environmental endocrine disruptors in terms of the maintenance of stemness and pluripotency.     

Treatment of phenol in synthetic saline wastewater by solvent extraction and two-phase membrane biodegradation

Phenol in synthetic saline (100gL(-1) NaCl) and acidic (pH 3) wastewater was treated by a hybrid solvent extraction and two-phase membrane biodegradation process at 30 degrees C. Kerosene was adopted to be the organic solvent because it was biocompatible and had a suitable partition coefficient for phenol. Phenol in water was first extracted by kerosene in a batch stirred vessel and the loaded solvent was passed through the lumen of a polyvinylidene fluoride (PVDF) hollow-fiber membrane contactor; in the meantime, Pseudomonas putida BCRC 14365 in mineral salt medium was flowed across the shell, to which tetrasodium phyophosphate (1gL(-1)) was added as a dispersing agent. The effect of the initial phenol level in wastewater (110-2400mgL(-1)) on phenol removal and cell growth was experimentally studied. At a cell concentration of 0.023gL(-1), it was shown that the removal of phenol from saline wastewater was more efficient at a level of 2000mgL(-1) when 0.02-m(2) membrane module was used. The effects of bigger membrane module size (0.19m(2) area) and higher initial cell concentration (0.092-0.23gL(-1)) on the performance of such a hybrid process for the treatment of higher-level phenol in saline wastewater was also evaluated and discussed.     

On mining multi-time-interval sequential patterns

Sequential pattern mining is essential in many applications, including computational biology, consumer behavior analysis, web log analysis, etc. Although sequential patterns can tell us what items are frequently to be purchased together and in what order, they cannot provide information about the time span between items for decision support. Previous studies dealing with this problem either set time constraints to restrict the patterns discovered or define time-intervals between two successive items to provide time information. Accordingly, the first approach falls short in providing clear time-interval information while the second cannot discover time-interval information between two non-successive items in a sequential pattern. To provide more time-related knowledge, we define a new variant of time-interval sequential patterns, called multi-time-interval sequential patterns, which can reveal the time-intervals between all pairs of items in a pattern. Accordingly, we develop two efficient algorithms, called the MI-Apriori and MI-PrefixSpan algorithms, to solve this problem. The experimental results show that the MI-PrefixSpan algorithm is faster than the MI-Apriori algorithm, but the MI-Apriori algorithm has better scalability in long sequence data.     

An adaptable and scalable group access control scheme for managing wireless sensor networks

Access control is a prime technology to prevent unauthorized access to private information, which is one of the essential issues appearing in secure group communication (SGC) of wireless sensor networks (WSNs). Many studies have made good progress on access control; however, their methods are inadequate to cope with this new issue for SGC-based WSNs since of their inflexibility, inefficiency, insecurity, or small-scale.