Search

International audience; Process mining techniques allow organizations to discover, monitor and improve their as-is processes by analyzing the process execution data, aka event data, recorded by their information systems. A recurrent task in process mining is querying. Querying allows users to get insights into specific executions of their processes and to retrieve relevant data. Existing process querying techniques require end users to be knowledgeable of the query language and the database schema. However, a key success factor for process analysis is to make querying accessible to business experts who may be inexperienced in database querying. This paper addresses this challenge by proposing a natural language interface (NLI) for querying event data. The interface allows users to formulate their questions in natural language and to automatically translate the questions into a structured query that can be executed over a database. We use graph based storage techniques, namely labeled property graphs, which allow to explicitly model event data relationships. As an executable query language, we use the Cypher language which is widely used for querying property graphs. The approach has been implemented and evaluated using a publicly available event log.

Global warming is becoming a paramount concern in the world. One way to decrease the effect of global warming is by decreasing carbon emission and using renewable energy. In particular, there are many works on using renewable energy technologies in mobile communication systems. In order to enable such technologies in mobile communication systems, we should be able to estimate the required energy. Most of research was focusing on techniques to be used to exploit renewable energy sources assuming that the required energy to run the base stations is known. Only few works were focusing on the estimation of the energy based on transmitted energy, and fewer relating the former to traffic. In this paper, we present a regression-based power consumption estimation method based on voice and data traffic provided by base stations with 2G and 3G capabilities. Our results show that the power consumption of different base stations as a function of the provided traffic can have different patterns. Furthermore, the same base stations can have different energy consumption models at different period of time. Therefore, we advocate the use of machine learning algorithms inside each base station to learn its specific pattern.

International audience

International audience; In the video analysis domain, automatic detection of actions performed in a recorded video represents an important scientific and industrial challenge. This paper presents a new method to approximate the boundaries of actions performed by a person while interacting with his environment (such as moving objects). This method relies on a Codebook quantization method to analyze the rough evolution of each pixel and then decide whether this evolution corresponds to an action or not; this decision is taken by an automated system. Statistics are then produced - at the scale of the whole frame - to estimate the start and the end of an action. According to our proposed evaluation protocol, this method produces interesting results on both real and simulated videos. This statistic-based protocol is discussed at the end of this paper. The interpretation of this evaluation protocol nominates this method to be a solid base to localize the exact boundaries of actions or - in the framework of this research activity - to associate prescriptive text with a visual content.

In this paper we propose a method to locate inloop repetitions in a video. An in-loop repetition consists in repeating the same action(s) many times consecutively. The proposed method adapts and uses the auto-correlation method YIN, originally proposed to find the fundamental frequency in audio signals. Based on this technique, we propose a method that generates a matrix where repetitions correspond to triangle-shaped zones of low values in this matrix (we called YIN-Matrix). Locating these triangles leads to locate video segments that enclose a repetition as well as to extract their parameters. In order to evaluate our method, we used a standard evaluation method that shows the error rates compared to ground-truth information. According to this evaluation method, our method shows promising results that nominate it to form a solid base for future works.

International audience; The decarbonization of the transportation sector needs a major rise in the electric vehicle (EV) market share in order to totally switch into electromobility. Boosting the electric vehicle market requires a cooperation between automotive industries by developing this technology especially batteries, charging infrastructure by installing more charging points especially fast ones and EV owners by giving them subsidies and offers. We collected data from different sources to analyze PEV sales in French departments and to know the reason that has the highest impact on the client's choice. Based on existing literature, we identified the most important factors and tried to build the French econometrics model using RStudio. Our model found that the vehicle price, autonomy, department's population density, local subsidies and fuel price to be significant and positively correlated to local PEV sales. However, charging infrastructure had negative impact and no significancy on the electromobility market. Results suggest boosting the study on a more detailed concept such as cities and suburbs as well as adding factors that reflect a department's and a client's characteristics in order to conclude with results that are more accurate.