These criticisms are valid for any kind of model development, and deserves an equally valid reply: the use of the ‘InnovationMining’ framework requires the same knowledge for both model building and result interpretations that other models demand. The innovationmining methodology should be regarded as an assistance to the product developer who wants to understand the interactions and relationships in a complex system from a ‘Systems Thinking’ perspective. Construction of the input data, especially the need to linking VOCs, EMs, subsystems and components, requires diligent thinking, as does any form of quantitative model. In addition, the mechanics of the innovationmining model is free from assumptions as the clustering technique requires no prejudice about the size and number of clusters. When used for revealing hidden relationships and identifying innovation opportunities, this gives a substantial advantage over the intuitive and qualitative approaches, because it provides insights on underlying relationships that may otherwise be ignored or unexplored. The resulting innovation opportunities from the case study which are the output of the innovationmining model can serve to evoke the product developer. He is free to reflect upon them, reject them, or modify them in any way.
Christensen  in his theory of disruptive innovation conducts a detailed study on the disk drive industry and explains patterns of innovation. He explains that in the disk drive industry, the first disk drive was developed by IBM which was a large in size and could store only five megabytes of data. Later it kept on improving and by 1976, disk drives were a billion dollar worth industry. The rate of industry growth continued and many companies that entered the market in the beginning failed to survive in the market by 1996. In his theory, he also explains that sustaining innovations are architectural changes that occur in the product, and they do not force companies to fall. In contrast to this, the disruptive innovations often cause companies to fall. One such example is that of the transition from fourteen inches to eight-inch disks. The companies which have been developing fourteen-inch drives did not choose to switch to eight-inch drives since their main customers were not them and there were only mainframe computers mainly in use then. With new applications like that of a mini computer, the smaller disk drives gained importance and eventually they completely replaced the fourteen-inch drive market. This led to the downfall of many fourteen-inch drive manufacturers .
Mining software repositories (MSR) is one among of the exciting and most rapid growing disciplines of software engineering. . Quite a lot codebase of software projects comprises of comments that record the deployment of the codebase, and it helps the contributors to comprehend the code structure, for modification and code reuse mainly. A number of experts have conducted studies expressing that code having comments message is less complicated to understand than code without having message i.e., comments , . Comments are the subsequent best used scripted thing for understanding the code, after the code . Additionally, a source code document is also imperative in maintenance and form a mainstay of the general documentation of the software projects. Contrary to exterior documentation, comments in the codebase are a straightforward mode for contributors to preserve related documents and keep the code consistent and up-to-date. Contributors generally admit that an inadequate document creates confusion , and studies mentioned in  state that poor documentation significantly have the least capacity of maintainability of the system. While contributors ordinarily concur on the interest of software manual , that putting the comments in code is usually ignored mainly because of strict release deadlines along with time constraints throughout the development. In the past few decades, software engineering strategies get continuously grew to bring cultural, social, technological, and organizational improvements. This improvement is a change in development strategies and methods from document driven to agile evolution that concentrates on developing the best software system instead of developing best documents . This progressive nature of changes results to scenarios wherein source code and comments usually the sole existing system artifacts having code layout, execution, and implementing the decision making. Studies and research have shown that the efficacious usages of the message in the form of comments can
is voided using cross validation with k folds and evaluated using the accuracy measures: precision and recall and roc. In this paper random tree classifier are given to better result as compared to decision stump . Seoung-hun Park and Young- gunk Ha is used imbalance technique and maps reduce algorithm .imbalance data means data that have a huge difference between the obverted sizes from one data set. So researcher are solved the problem to used sampling technique. There are two type sampling are: over sampling and under sampling .over sampling is to use all observation value in a big class and increase size of observation value in a small class and use this value. Under sampling are those who used lost data. Other problem are occur the researcher are data processing .in this case training set of data make multiple feature .it take time so much so that researcher are solve the problem in map reduce algorithm map reduce algorithm are used to big processing technique.
The website-based indicator of R&D is also not significantly correlated with patents and publications. Patents and publications show important formal results of R&D activities but not the other complementary processes that bring the results of formal R&D activities to the marketplace (and which are often not well-captured by conventional data sources). It is well documented that there are a number of factors that influence patenting and publishing behavior and firms may decide not to patent or publish the outputs of their R&D activity. Firms may also choose not to place all details of their R&D activities on public websites, but they can generally describe what they are doing for the benefit of customers (real and potential) without revealing extensive technical details to their competitors. This helps to explain the lack of correlation between the website-based indicator of R&D activity and patent and publication based R&D output. The lack of correlation between different conventional indicators of R&D, their relative strengths and weaknesses, and the particular biases associated with them is already well documented in the literature (see, for example, Kleinknecht et al. 2002). The repeated conclusion in many studies is to recommend use of indicators that are appropriate to the purpose, context and particular issues being inves- tigated. Our conclusions augment the previous literature by explicating a framework of relative strengths and weaknesses of website data. This can help in making choices about which data sources and approaches to use in a particular study.
54 O’Keele and Preece (1996) noted that the verification of a system can be achieved using three measures: conflict, redundancy, and deficiency. Conflict refers to the ability of the system to arrive at logically inconsistent conclusions from consistent input; redundancy refers to the presence within the system of logically unnecessary structures that never affect the relationship between the input and output of the system; and deficiency refers to the absence of structures that should be present, logically, for the system to arrive at conclusions for all valid input cases. They worked on verifying their system attempting to detect anomalies. Thus, an anomaly could indicate any of the above measures. Anomaly detection is focused on the usage of rules. In terms of validation, O’Keele and Preece presented a list of methods used for calibration of knowledge-based systems (rules, heuristic, case testing) and also suggested that a strategy should be implemented for validation and verification of a knowledge-based system. They suggested the following guidelines for the development of a strategy:
We have analysed the scientific literature that investigates which factors are important in innovation processes. Many of these papers classify important indicators in relatively similar dimensions that have been used in the Innovation Radar. Balachandra and Friar (1997) proposes four major categories on market, technology, environment, and organisational related characteristics. These categories have been widely recognised and adopted by many scholars in the field of technology commercialisation of R&D projects (Astebro, 2004; Linton et al., 2002). Alternatively, Heslop et al. (2001) use factor analyses to group more than fifty variables related to the technology commercialisation process into four dimensions of market readiness, technology readiness, commercial readiness, and management readiness. However, there is no clear evidence of which dimension is more important in the innovation process.
Data mining tasks can be classified to tasks of description and prediction. While description aims at finding human- interpretable patterns and associations, after considering the data as a whole and constructing a model prediction seeks to foretell some response of interest. Although the goals of description and prediction may overlap, the main distinction is that prediction requires the data to include a special response variable .The models generated by some prediction methods may point out some interesting patterns. The goal of predictive data mining in clinical medicine is to construct a predictive model that is sound, makes reliable predictions and helps physicians improve their prognosis, diagnosis or treatment planning procedures.
The Parmenides project developed a text mining application applied in three different domains exemplified by case studies for the three user partners in the project. During the lifetime of the project (and in parallel with the development of the system itself) an evaluation framework was developed by the authors in conjunction with the users, and was eventually applied to the system. The object of the exercise was two-fold: firstly to develop and perform a complete user-centered evaluation of the system to assess how well it answered the users' requirements and, secondly, to develop a general framework which could be applied in the context of other users' requirements and (with some modification) to similar systems. In this paper we describe not only the framework but the process of building and parameterising the quality model for each case study and, perhaps most interestingly, the way in which the quality model and users' requirements and expectations evolved over time.
In this paper, we propose a framework for APP discovery for a given user based on APP relation mining. First, we construct personalized trust community (PTC) with complex network techniques from APP-user related data. Then the APP is discovered through APP set matching between given user and similarity user in PTC. We plan to experiment our framework with the data in APP store. Due to the timeliness of APP, we plan to crawl the data in one year. All the APP related reviews will be collected for experimentation, although most reviews have no connection with the given user. PTC can also be used for malicious APPs filtering, preference mining and so on.
From the above mentioned, it appears that intrusion detection is a relevant challenge in information system security. This paper presents a model designed to detect intrusion by workflow mining that permits to analyze event logs presenting events related to resources of the considered system. This approach helps to monitor resources directly and then, detect as intrusive, all actions that violates the security policy built around the rights and permissions defined by the managers of information systems for the manipulation of resources. Moreover, this model provides a solution for the problem of the high rate of false alerts because intrusion do not use training data but the quality of rules that represents the security policy. One of the major challenges to handle here is the management of a large volume of data present in event logs. Another challenge is the manner to find a canonical set of rules for the security policy such that, all the rules can be related to one or many rules present in the set of canonical rules. After experimentation, we have compared our solution to three others solutions and we have found that our prototype deployed on windows system to monitor a local disc C has a better accuracy for detection using workflow mining. The problem of false alerts is managed in another angle and has a better solution. But, this solution implies a good strategy for the construction of security policy. Future works can tackle the challenge of integration of big data techniques to improve intrusion detection within an information system using workflow mining. Moreover, the definition of an automatic model for security policy definition appears like another relevant issue.
gives a lot of recognition to science, technology and innovation (ST&I) as the main way to be able to achieve advancement in economically, politically and environmentally. ST&I and other strategies are crucial for the improvement of technology, and innovation. For Kenya to be able to harness science, technology and innovation, the government has to put in place effective regulatory policies. There has been the growth of ICT in Kenya which can be shown by the number of telephone lines, internet service providers (ISPs), and the rising internet users, broadcasting stations and liberalization of the mobile cellular market and there are now two cellular mobile operators. Over 70% of the people in the country have access to television, and over 95% have access to radio services. The government of Kenya in the year 1997 released a policy guiding postal and telecommunication sector that allowed competition and the Kenya Communication Act was enacted in 1998 (Mambi, 2010).
and devouring the fruits of innovations (Christensen, 1997). Well established firms keep seeking ways of innovations in order to remain at the upper-stream of supply-chain and value-chain, leaving those startups and followers at the mid- and lower- stream of supply-chain and value- chain. Such a ‗cat and mouse‘ kind of relationship between innovators and imitators has been used to outline the dominant mainstream of innovation literature, which has not changed much during the past decade or so, except the continuously extended discussions on either ‗how innovators can keep leading‘, or, ‗how imitators can catch up‘. Such a historically inherited and stereotyped mindset seems to become the ruling principle of academia game in exploring and disclosing the dynamism and mechanism of profit maximization. Facing such an embarrassed academic challenge, there comes a group of scholars advocating the legitimacy of imitation, arguing that, both innovation and imitation are civilized ways of profit-making, and the two compete against each other based upon their pathways of gaining and sustaining their respective capabilities and advantages, as well as their respective speed of delivering products/services to market places (Anthony et al., 2008; Assink, 2006; Chittoor et al., 2009; Cooper & Edgett, 2008; Husig et al., 2005; Kale & Little, 2007; Kim, 1997; Lawson & Samson, 2001; Pil & Cohen, 2006; Shenkar, 2010; Stieglitz & Heine, 2007; Zeng, 2007; Zeschky et al., 2011; Zhao, 2008; 2012; Zhao & Zhang, 2016; 2017).
In this context a number of studies have already investigated the potential viability of new land-use systems within New Zealand (for example see CDC, 2015; Coriolis, 2012; Boyd, 2016; Bryan, 2015). These studies have identified, for example, dairy sheep, dairy goats, hazelnuts and Manuka honey as systems with the potential for significant expansion. However, in addition to identifying potential NGS there is the need to take into account the context in which any system is likely to be implemented. For example, land-owners’ goals and objectives clearly vary (Pannell et al., 2012) and this directly influences whether or not new systems are adopted. In addition, there is spatial variation in such factors as climate, topography, regulatory pressures (for example from nutrient emission limits) and production opportunities (for example from new irrigation schemes). The requirements from novel systems and their likelihood of being adopted will, therefore, vary considerably across New Zealand. This means that any framework to identify potential NGS needs to take into account the situation of the individual land-manager. In the following sections we describe the development of such a framework and test its usefulness using a simple example involving two land managers and one possible NGS – dairy sheep.
In comparison, the situation with regard to the previ- ous Android example is different. This radical inno- vation required a tremendous modification of the business model and its components since the value proposition, value constellation, and the entire value creation process became subject to substantial change. From a business model component perspective, devel- oping and marketing Android had a massive impact on the sub-models. The network model, for example, had to be modified since new business partners needed to become part of the development and marketing pro- cess (e.g., open developers, network carriers, and hand- set manufacturers). Moreover, developing platform software requires new competencies and additional resources, which calls for an adaption of the competen- cies and resources model. Similarly, the new product/ service offer demands an entirely distinctive manufac- turing and revenue model. These examples underline the importance of being aware of the BMI Intensity since a moderate innovation should carry less risk and take less effort compared to a more intense innovation. BMI outcome/impact
of Process-Aware Information Systems and as a lecturer at the Software Development Management Department of the Higher School of Economics in Moscow. In 2013 he had an internship at the Technical University of Eindhoven, the Netherlands. His publications cover a number of research interests such as digital HF electronics, software verification and architectures, and process mining. The following is one of his publications: Mitsyuk, A., Nkova, A. K., Shershakov, S., & van der Aalst, W. (2014). Using process mining for the analysis of an e-trade system: A case study. Business Informatics, 3(29).
Narrow reef mechanization is critical and one only needs to review the six SAIMM International Platinum Conferences from 2004 until 2014 to understand the volume of work that has gone into introducing mechanization to the southern African platinum industry (Figure 7) . One cannot raise the issue of mechanization in the platinum industry without discussing Lonmin and its 2004 vision to achieve 50% of reef production and 100% for development metres by 2010 using mechanized mining methods (Webber et. al., 2010). The actual level of mechanization achieved in 2010 was only 20%. The Lonmin mechanization programme failed to meet its target with equipment underperforming, difficulties in regard to labour and supervision, and mining dilution higher than planned due to the size of on-reef developmnt and the inability of the extra-low profile (XLP) equipment to handle rolls in the reef. Thus, Lonmin changed its strategy, deciding that mechanized mining utilizing the XLP equipment needed to be proved on a smaller scale before implementing it mine-wide.
We propose a collaborative requirement-miningframework that addresses the problem of an OEM receiving a request for proposal consisting in a StRS and the referenced applicable documents, all of them prescribing a massive set of requirements that cannot be implemented within cost and schedule constraints. The mission of the framework is to support the business analysts willing to distil a large set of requirements, that is, to create an optimised Sytem Require- ments Specification (SyRS) that is ready to be managed in configuration thanks to requirements management tools. By using the “distil” we mean to get and show only the most important part of a large set of requirements. We qualify the SyRS as optimised because it is the outcome of the distilling process that results in a subset of key requirements. The key requirements gather:
straints that can be added automatically by the execution mechanism. A redun- dant constraint is already implied by the model – it does not express an actual restriction of the solution space – but it can potentially improve solver per- formance, e.g. by contributing additional constraint propagation. A predicate implementing a redundant constraint for minimum frequent itemset mining is shown in Listing 7. It uses the insight that if an itemset must be frequent, then each item must be frequent as well; hence, items that appear in too few transac- tions can be removed without searching over them. This can be encoded with a constraint that performs look-ahead on the items (Listing 7, line 6). See  for a more detailed study of this constraint. Another type of redundant information available in the library is a search annotation (Listing 4, line 15). This is an an- notation that can be added to the search keyword, and that specifies the order in which to search over the variables. An example of a search order that has been shown to work well for itemset mining is occurrence . We also added the enumerate search annotation to differentiate, in the model, between satisfaction (one solution) and enumeration (all solutions) problems. The last annotation is the query keyword, which can be added to a variable declaration, for example array  of set : TDB :: query(”mydb.sql”, ”SELECT tid,item FROM purchases”); . The execution mechanism will automatically typecheck the expression, execute the query and add the data as an assignment to that variable. In this way, one can directly load data from a database, as is common in data mining.
This document presents findings from the first phase of the assessment of the ‘Life Essentials Assessment Framework’ (‘Leaf’) questionnaire, a six questions, interviewer- administered questionnaire devised by Age UK Wakefield District to enable effective evaluation of vulnerable adults’ needs and to help establish the effectiveness of service provision. The assessment aimed to investigate the validity, reliability, and capacity to measure change of the ‘Leaf’ questionnaire in two phases: