Abstract
Extraction and normalization of temporal expressions are essential for many NLP tasks. While a considerable effort has been put on this task over the last few years, most of the research has been conducted on the English domain, and only a few works have been developed on other languages. In this paper, we present ParsTime, a tagger for temporal expressions in Persian (Farsi) documents. ParsTime is a rule-based system that extracts and normalizes Persian temporal expressions according to the TIMEX3 annotation standard. Our experimental results show that ParsTime can identify temporal expressions in Persian texts with an F1-score 0.89. As an additional contribution we make available our code to the research community.

Abstract

Abstract
The development of information retrieval algorithms and temporal information retrieval ones has been extensively carried out over the last few years. While several studies have been conducted, most of these researches relate to English, leading to a lack of knowledge in several other important languages. This includes the Persian one. In this work, we aim to shorten this gap by contributing, disseminating and enlarging the knowledge we have on temporal information retrieval aspects in Persian, which is one of the dominant languages in the Middle East, widely spoken in several countries. To achieve this objective, we propose to understand the use of temporal expressions on a large-scale Persian search engine query log consisting of 27M queries. In particular, we focus on explicit (e.g., June 2017) and relative temporal expressions (e.g., tomorrow) and try to understand (1) how often temporal expressions are used in web queries; (2) which type of temporal expressions (Date, Time, Duration and Set) are commonly used; (3) to which time (past, current or future) do temporal expressions mostly refer to; (4) to which category they often belong; (5) how often do user's reformulate their queries by adding temporal expressions; and (6) how using temporal expressions affects user's satisfaction. We believe that answering these questions may be beneficial for a large number of tasks including, user's behavior understanding and search engines' improvement effectiveness.

Abstract
Over the last few years, an increasing number of user's and enterprises on the internet has generated a global marketplace for both employers and job seekers. Despite the fact that online job search is now more preferable than traditional methods - leading to better matches between the job seekers and the employer's intents - there is still little insight into how online job searches are different from general web searches. In this paper, we explore the different characteristics of online job search and their differences with general searches, by leveraging search engine query logs. Our experimental results show that job searches have specific attributes which can be used by search engines to increase the quality of the search results.

Abstract
Human language constantly evolves due to the changing world and the need for easier forms of expression and communication. Our knowledge of language evolution is however still fragmentary despite significant interest of both researchers as well as wider public in the evolution of language. In this paper, we present an interactive framework that permits users study the evolution of words and concepts. The system we propose offers a rich online interface allowing arbitrary queries and complex analytics over large scale historical textual data, letting users investigate changes in meaning, context and word relationships across time.

Abstract
The permutation flow shop scheduling problem is one of the most studied operations research related problems. Literally, hundreds of exact and approximate algorithms have been proposed to optimise several objective functions. In this paper we address the total tardiness criterion, which is aimed towards the satisfaction of customers in a make-to-order scenario. Although several approximate algorithms have been proposed for this problem in the literature, recent contributions for related problems suggest that there is room for improving the current available algorithms. Thus, our contribution is twofold: First, we propose a fast beam-search-based constructive heuristic that estimates the quality of partial sequences without a complete evaluation of their objective function. Second, using this constructive heuristic as initial solution, eight variations of an iterated-greedy-based algorithm are proposed. A comprehensive computational evaluation is performed to establish the efficiency of our proposals against the existing heuristics and metaheuristics for the problem.