Search results

To accurately forecast logistics freight volume plays a vital part in rational planning formulation for a country. The purpose of this paper is to contribute to developing a novel combination forecasting model to predict China’s logistics freight volume, in which an improved PSO-BP neural network is proposed to determine the combination weights.

Since BP neural network has the ability of learning, storing, and recalling information that given by individual forecasting models, it is effective in determining the combination weights of combination forecasting model. First, an improved PSO based on simulated annealing method and space-time adjustment strategy (SAPSO) is proposed to solve out the connection weights of BP neural network, which overcomes the problems of local optimum traps, low precision and poor convergence during BP neural network training process. Then, a novel combination forecast model based on SAPSO-BP neural network is established.

Simulation tests prove that the proposed SAPSO has better convergence performance and more stability. At the same time, combination forecasting models based on three types of BP neural networks are developed, which rank as SAPSO-BP, PSO-BP and BP in accordance with mean absolute percentage error (MAPE) and convergent speed. Also the proposed combination model based on SAPSO-BP shows its superiority, compared with some other combination weight assignment methods.

SAPSO-BP neural network is an original contribution to the combination weight assignment methods of combination forecasting model, which has better convergence performance and more stability.

The helpfulness vote is a type of aggregate user representation that, by measuring the quality of an online review based on certain criteria, can allow readers to find helpful reviews more quickly. Although widely applied in practice, the effectiveness of the voting mechanism is unsatisfactory. This paper uses the heuristic–systematic model and the theory of dynamics of reviews to shed light on the effect of various information cues (product ratings, word count and product attributes in the textual content of reviews) on online reviews’ aggregative voting process. It proposes a conceptual model of seven empirically tested hypotheses.

A dataset of user-generated online hotel reviews (n = 6,099) was automatically extracted from Ctrip.com. In order to measure the variable of product attributes as a systematic cue, the paper uses Chinese word segmentation, a part-of-speech tag and word frequency statistics to analyze online textual content. To verify the seven hypotheses, SPSS 17.0 was used to perform multiple linear regression.

The results show that the aggregative process of helpfulness voting can be divided into two stages, initial and cumulative voting, depending on whether voting is affected by the previous votes. Heuristic (product ratings, word count) and systematic cues (product attributes in the textual content) respectively exert a greater impact on the two stages. Furthermore, the interaction of heuristic and systematic cues plays an important role in both stages, with a stronger impact on the cumulative voting stage and a weaker one on the initial stage.

This paper’s findings can be used to explore improvements to helpfulness voting by aligning it with an individual’s information process strategy, such as by providing more explicating heuristic cues, developing different methods of presenting relevant cues to promote the voting decision at different stages, and specifying the cognitive mechanisms when designing the functions and features of helpfulness voting.

This study explores the aggregative process of helpfulness votes, drawing on the study of the dynamics of online reviews for the first time. It also contributes to the understanding of the influence of various information cues on the process from an information process perspective.