FATMAP is the outdoor adventure platform for web and mobile. Wherever your inspiration takes you, you'll be ready for any adventure. From those late night planning sessions to those early morning summits.

FATMAP

The future is always full of uncertainty, and with the explosive growth of remote work due to the COVID-19 pandemic, many are wondering what will happen to the workplace as the world slowly reopens. One thing that remains clear is that the pandemic has changed some things in our world permanently. Remote work is very likely one of those things. Although some workers will have to return to in-person work as restrictions are lifted, we should continue to see an overall steady rise in remote work. Lockdowns and quarantines forced many businesses to figure out a way to work remotely, and now that they’ve seen the benefits and overcome many of the challenges, there’s simply no turning back. Here are just some reasons why remote work is not going away. 1. The Pandemic is Not Dying Down Thanks to the available vaccines, hospitalizations, and deaths from COVID-19 are down in many parts of the world, but the virus is still spreading in other places. Many that COVID-19 will become endemic and seasonal like the flu. It will hopefully become less deadly, but at this point, it is hard to say. Bringing everyone back to office space only to have to go virtual again at some point in the future will cause unnecessary stress and waste resources. scientists speculate Even if COVID-19 disappears completely, there is no reason to think that there won’t be another pandemic from a new virus in the near future. that due to climate change and further encroachment into animal habitats, there may be more new viruses that mutate and “jump” from animals to humans. Hopefully, this doesn’t happen, but the future is uncertain. Scientists worry What we do know is that remote work is better for employees’ physical health in general. With more remote work, there will likely be less spread of other viruses and illnesses such as the common cold and the flu as we saw in 2020. 2. Businesses are Realizing the Value of Remote Work There are numerous benefits to businesses for having a remote team. Before the pandemic, many businesses had not yet tried offering remote work, likely because they felt that productivity would decrease, workers would be unmotivated and distracted and communication would be a struggle. However, now that many companies have been forced to try out remote work due to lockdowns and regulations, they have learned that many of their fears about remote work were unfounded. Further, businesses have had a chance to see the benefits of remote work firsthand. It Saves Money It is estimated that a typical employer can save an average of remote worker per year. With a full-time remote team, businesses could save up to $22,000 per worker, per year, in addition to saving on overhead such as an office. $11,000 per half-time Employees save money as well by not having to spend as much on coffee, lunch, a professional wardrobe, and gas. A study from estimates that workers can save as much as $4,000 per year by working remotely. Flexjobs Easier Recruitment and Better Retention The overwhelming majority of workers prefer doing remote work to commuting into a physical office. Over 80% of workers surveyed by reported that working remotely would help them to be happier, feel more trusted, and be better able to handle work-life conflict. Owl Labs Offering remote work is a huge perk that will help employees stick around longer and will attract new talent. Increased Productivity A study done by found that most workers either kept the same level of productivity after a transition to remote work or increased their productivity. Without a commute or in-person meetings to attend, many workers reported that they were able to get more done. Great Place to Work Less Sick and Personal Days Used As mentioned above, remote work is generally better for the physical health of employees, even when not in the midst of a pandemic. Additionally, seasonal viruses and illnesses won’t spread through a remote team the way they would through an in-person team working together. Even though everyone on a team might get sick, it is very unlikely that they will all get sick at the same time. A found that remote workers take an average of 2.4 sick days per year, in comparison to the 2.6 taken by those working from company premises. study done by CartridgePeople.com With flexible work-from-home schedules and better work-life balance, it is likely that remote workers need to take fewer personal days as well. 3. Workers Are Unable and/or Unwilling to Return In-Person Since the start of the pandemic, many workers have left large cities and moved away from their offices. Others may have childcare issues due to adjustments made during lockdowns. Still, others have simply gotten used to all the benefits of remote work and are unwilling to return to an in-person workspace. According to , many people have already left jobs over the fact that they were requiring a return to the office. Managers of in-person teams also reported that it was “exceedingly difficult” to find candidates, especially quality ones, to fill an in-person-only position. research done by Slate 4. Global Teams Are On the Rise Worldwide employment by multinational companies headquartered in the U.S. and abroad has been and this is a trend that will likely continue. Diversity of thinking, creativity, and access to talent from anywhere are the top benefits noted by companies with global teams. Remote work allows companies to easily hire someone from anywhere in the world without having to ask them to move. Because of this, as the number of global teams increases, so will remote work. increasing 5. Better for the Environment Companies of all sizes are beginning to realize the importance of taking steps to help the environment. As things like climate change, and air pollution become more urgent problems. We are seeing large brands like taking big green initiatives. Switching to a remote team is another easy way that companies can help the environment by reducing travel significantly for their workers. Ikea, Apple, and Nike According to Fatmap, In 2017, transport was the largest contributor to greenhouse gas (GHG) emissions in the UK. Most employees cut their overall travel time down significantly when they switch to remote work. We saw the positive effect of this during the height of the pandemic when there was less congestion, cleaner air, and other . positive impacts on the environment Conclusion: It’s Time To Embrace Remote Work Although remote work was not in the game plan for many companies, it is clear now that embracing remote work is a smart choice. There are numerous benefits across the board and most of the challenges are easily overcome with the right cloud communication system and tech stack. In the near future, it is likely that working at a physical office will be the exception, not the rule.

5 Reasons Why Remote Work is Not Going Away

DISCLAIMER: The code included in this tutorial belongs to TJ Tsai, the creator of Bootleg Score. This tutorial is designed to facilitate the initial approach for new researchers. If you have any questions, do not hesitate to contact him: ttsai@g.hmc.edu. Without Optical Music Recognition (OMR), computers can not be able to interpret a music score, and a lot of tasks related with Music Information Retrieval can not be achieved. In simple words, OMR transforms the data from music scores into a format that machines can understand, with some famous music representations such as MIDI, MusicXML or PianoRoll. In this short tutorial, I aim to introduce you to Bootleg Score, a mid-level music representation that facilitates music sheet processing without requiring significant computational power or extensive libraries and dependencies. Since I have been working with this representation lately, I think its possibilities should be known by more people. This representation was created by TJ Tsai and his lab from Harvey Mudd College, which seeks to represent the position of filled notes on a staff using zeros and ones, explicitly encoding the rules of Western Music Notation for piano sheet music. Some of their articles showing applications and results obtained by performing tasks with Bootleg Score representation include the following: MIDI Passage Retrieval Using Cell Phone Pictures Of Sheet Music
Composer Style Classification Of Piano Sheet Music Images Using Language Model Pretraining
Camera-Based Piano Sheet Music Identification
MIDI-Sheet Music Alignment Using Bootleg Score Synthesis In order to facilitate the first approach to this representation, all the necessary files to get the Bootleg Score representation of a PDF sheet music are available in the following github repository: How To Use Bootleg Score INSTALLMENT All requirements are found in the requirements.txt file of the repository. I recommend working in a Linux system such as Ubuntu. Once you create a Conda Environment, installing all the dependencies is as easy as writing the following command in the terminal. pip install -r requirements.txt Some of the most important libraries are OpenCV, frequently used for computer vision, and Scikit-Learn, used for data analysis. HOW DOES BOOTLEG SCORE WORKS To get this low resource optical recognition system, it is necessary to focus on three things on the sheet music: filled noteheads, staff lines and bar lines. All the functions mentioned in the next paragraphs are included in the ExtractBootlegFeatures1.py file from the repository. NOTEHEAD DETECTION The filled noteheads are selected by using a simple blob detector and estimating the noteheads that fill in a notehead template. The function adaptiveNoteheadDetect() performs the adaptive detection of noteheads from our image. Finally, with the notes list data (nhlocs), we can obtain the required information of all the noteheads detected. Another function takes a list of bounding boxes representing noteheads, calculates the central coordinates of each one of them, and provides an average estimation of the width and length of the noteheads in that list. STAFF LINE DETECTION The staff lines location and spacing are estimated using every detected notehead from the previous step. The function isolateStaffLines() uses morphological operations to isolate the staff lines in an image. First, the horizontal lines are identified and then the note bars are removed to obtain only the staff lines. With the getEstStaffLineLocs() function, we have a list of predictions where each prediction is a tuple containing the initial and final position of a staff line, the associated column and row, and the corresponding index in the featmap. The next step consists of assigning the noteheads to the staff that corresponds to them.The final result will be similar to what is shown in the following image. BAR LINE DETECTION The Bar Line detection helps to cluster the staff line into grand staff systems. It is important to remember that the Bootleg Score representation is taught to be used with Piano Sheet Music. Bounding boxes are predicted around the bar lines in the image. Bar lines are necessary to properly group the staff lines into clusters of grand staff systems, each one containing a staff corresponding to the right hand and another corresponding to the left hand. BOOTLEG SCORE GENERATION Using the information from the noteheads, staff lines, and bar lines, the Bootleg Score is generated as an array of 0’s and 1’s. The vertical axis represents the positions of the right and left hand staves, while the notes are located along the horizontal axis. For each single bootleg line, collapseSimultaneousEvents() groups the noteheads that are close to each other in the same column, assuming that these notes sound simultaneously. The function constructBootlegScore encodes the previous information into the Bootleg Score representation. The Bootleg representation is a matrix with dimensions of 62×N, where 62 represents the possible staff positions and N represents the number of simultaneous events. An example of how a Bootleg Score can look is shown in the next image: Once the final representation is done, it is time for you to try different tasks involving music information retrieval taking advantage of the possibilities that a low resource representation like this one provides to users. A sheet music PDF is included in the repository for testing. Once everything is correctly configured, the only thing you need to do is execute the Test_Bootleg.py file. Do not hesitate to take a deep look at the code; it is a good opportunity to learn new methods for analyzing sheet music, and I found it very funny.. CONCLUSIONS In conclusion, TJ Tsai’s Bootleg Score representation represents a significant progress in Optical Music Recognition, where computational costs and extensive processing times have been a problem for many years. The articles mentioned in previous sections demonstrate that using the Bootleg Score can achieve good results in various tasks, so I encourage you to try it and give it an opportunity. DISCLAIMER: The code included in this tutorial belongs to TJ Tsai, the creator of Bootleg Score. This tutorial is designed to facilitate the initial approach for new researchers. If you have any questions, do not hesitate to contact him: ttsai@g.hmc.edu. DISCLAIMER : The code included in this tutorial belongs to TJ Tsai, the creator of Bootleg Score. This tutorial is designed to facilitate the initial approach for new researchers. If you have any questions, do not hesitate to contact him: ttsai@g.hmc.edu . DISCLAIMER DISCLAIMER ttsai@g.hmc.edu ttsai@g.hmc.edu Without Optical Music Recognition (OMR), computers can not be able to interpret a music score, and a lot of tasks related with Music Information Retrieval can not be achieved. In simple words, OMR transforms the data from music scores into a format that machines can understand, with some famous music representations such as MIDI, MusicXML or PianoRoll. In this short tutorial, I aim to introduce you to Bootleg Score, a mid-level music representation that facilitates music sheet processing without requiring significant computational power or extensive libraries and dependencies. Since I have been working with this representation lately, I think its possibilities should be known by more people. This representation was created by TJ Tsai and his lab from Harvey Mudd College , which seeks to represent the position of filled notes on a staff using zeros and ones, explicitly encoding the rules of Western Music Notation for piano sheet music. Some of their articles showing applications and results obtained by performing tasks with Bootleg Score representation include the following: Harvey Mudd College Harvey Mudd College MIDI Passage Retrieval Using Cell Phone Pictures Of Sheet Music Composer Style Classification Of Piano Sheet Music Images Using Language Model Pretraining Camera-Based Piano Sheet Music Identification MIDI-Sheet Music Alignment Using Bootleg Score Synthesis MIDI Passage Retrieval Using Cell Phone Pictures Of Sheet Music MIDI Passage Retrieval Using Cell Phone Pictures Of Sheet Music MIDI Passage Retrieval Using Cell Phone Pictures Of Sheet Music MIDI Passage Retrieval Using Cell Phone Pictures Of Sheet Music Composer Style Classification Of Piano Sheet Music Images Using Language Model Pretraining Composer Style Classification Of Piano Sheet Music Images Using Language Model Pretraining Composer Style Classification Of Piano Sheet Music Images Using Language Model Pretraining Composer Style Classification Of Piano Sheet Music Images Using Language Model Pretraining Camera-Based Piano Sheet Music Identification Camera-Based Piano Sheet Music Identification Camera-Based Piano Sheet Music Identification Camera-Based Piano Sheet Music Identification MIDI-Sheet Music Alignment Using Bootleg Score Synthesis MIDI-Sheet Music Alignment Using Bootleg Score Synthesis MIDI-Sheet Music Alignment Using Bootleg Score Synthesis MIDI-Sheet Music Alignment Using Bootleg Score Synthesis In order to facilitate the first approach to this representation, all the necessary files to get the Bootleg Score representation of a PDF sheet music are available in the following github repository: How To Use Bootleg Score How To Use Bootleg Score How To Use Bootleg Score How To Use Bootleg Score INSTALLMENT INSTALLMENT All requirements are found in the requirements.txt file of the repository. I recommend working in a Linux system such as Ubuntu. Once you create a Conda Environment, installing all the dependencies is as easy as writing the following command in the terminal. requirements.txt pip install -r requirements.txt pip install -r requirements.txt Some of the most important libraries are OpenCV , frequently used for computer vision, and Scikit-Learn , used for data analysis. OpenCV OpenCV Scikit-Learn Scikit-Learn HOW DOES BOOTLEG SCORE WORKS HOW DOES BOOTLEG SCORE WORKS To get this low resource optical recognition system, it is necessary to focus on three things on the sheet music: filled noteheads, staff lines and bar lines. All the functions mentioned in the next paragraphs are included in the ExtractBootlegFeatures1.py file from the repository. NOTEHEAD DETECTION NOTEHEAD DETECTION The filled noteheads are selected by using a simple blob detector and estimating the noteheads that fill in a notehead template. The function adaptiveNoteheadDetect() performs the adaptive detection of noteheads from our image. Finally, with the notes list data (nhlocs), we can obtain the required information of all the noteheads detected. Another function takes a list of bounding boxes representing noteheads, calculates the central coordinates of each one of them, and provides an average estimation of the width and length of the noteheads in that list. STAFF LINE DETECTION STAFF LINE DETECTION The staff lines location and spacing are estimated using every detected notehead from the previous step. The function isolateStaffLines() uses morphological operations to isolate the staff lines in an image. First, the horizontal lines are identified and then the note bars are removed to obtain only the staff lines. With the getEstStaffLineLocs() function, we have a list of predictions where each prediction is a tuple containing the initial and final position of a staff line, the associated column and row, and the corresponding index in the featmap. The next step consists of assigning the noteheads to the staff that corresponds to them.The final result will be similar to what is shown in the following image. BAR LINE DETECTION BAR LINE DETECTION The Bar Line detection helps to cluster the staff line into grand staff systems. It is important to remember that the Bootleg Score representation is taught to be used with Piano Sheet Music. Bounding boxes are predicted around the bar lines in the image. Bar lines are necessary to properly group the staff lines into clusters of grand staff systems, each one containing a staff corresponding to the right hand and another corresponding to the left hand. BOOTLEG SCORE GENERATION BOOTLEG SCORE GENERATION Using the information from the noteheads, staff lines, and bar lines, the Bootleg Score is generated as an array of 0’s and 1’s. The vertical axis represents the positions of the right and left hand staves, while the notes are located along the horizontal axis. For each single bootleg line, collapseSimultaneousEvents() groups the noteheads that are close to each other in the same column, assuming that these notes sound simultaneously. The function constructBootlegScore encodes the previous information into the Bootleg Score representation. The Bootleg representation is a matrix with dimensions of 62×N, where 62 represents the possible staff positions and N represents the number of simultaneous events. An example of how a Bootleg Score can look is shown in the next image: Once the final representation is done, it is time for you to try different tasks involving music information retrieval taking advantage of the possibilities that a low resource representation like this one provides to users. A sheet music PDF is included in the repository for testing. Once everything is correctly configured, the only thing you need to do is execute the Test_Bootleg.py file. Do not hesitate to take a deep look at the code; it is a good opportunity to learn new methods for analyzing sheet music, and I found it very funny.. CONCLUSIONS CONCLUSIONS In conclusion, TJ Tsai’s Bootleg Score representation represents a significant progress in Optical Music Recognition, where computational costs and extensive processing times have been a problem for many years. The articles mentioned in previous sections demonstrate that using the Bootleg Score can achieve good results in various tasks, so I encourage you to try it and give it an opportunity.

How to Use Bootleg Score: A Low Resource Optical Music Recognition System

Comparison of Machine Learning Methods: Approach

Comparison of Machine Learning Methods: Conclusions and Future Work, and References

Authors:
(1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr};
(2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Table of Links Abstract and Introduction
Background
Approach
Experiments and Results
Discussion and Qualitative Analysis
Related Work
Conclusions and Future Work, and References 2 Background To understand our approach, it is essential to have a solid understanding of the structure of software issues in Issue Tracking Systems (ITS). 2.1 Software Issues in ITS This section explores software issues within ITS, covering their anatomy, life cycle, and significance in software projects. Issue Tracking Systems (ITS) An issue tracking system (ITS) is a software application that manages and maintains lists of issues. It provides a database of issue reports for a software project. Members of the software development team stay aligned and collaborate faster with the help of these tools. Users can log in with a password and post a new issue report or comment on an already-posted issue. There are various issue-tracking tools that are used in software development projects, such as JIRA, Bugzilla, and Azure DevOps. Although there are some differences among these tools, they all share a similar basic structure. We developed our approach with a data set of a company that uses JIRA software. Anatomy of an Issue The issue report in ITS contains various fields, each of which contains a different piece of information as shown in Figure 1. The following details [48] could typically be included in the generic process of reporting an issue: – Summary: Title of the issue. – Description: Issue details including its cause, location, and timing. – Version: Version of the project to which the issue belongs. – Component: The predetermined component on which the issue depends. – Attachment: Attaches such as pictures, videos, and documents uploaded to the issue. – Priority: Urgency level of the issue. – Severity: Frequent occurrence and systemic effects. – Status: Current status of the issue. – Reporter: Person who creates the issue. – Assignee: Person to whom the issue is assigned. – Revision History: Historical changes of the issue. – Estimated time: Estimated time spent to develop or solve the issue. – Comments: Additional details that can be used to understand the issue. The summary, description, and comments all contain textual details about the issue. In our methodology, we extract numerical features using the language structure of summary and description. Version is the current version of the project which differs with each release. The issue can have a predefined tag or component added to it. (e.g. project v1.1.0) Users can upload files to help others understand the issue, an attachment refers to it. Priority, severity, and status are the categorical features of the issue. Priority is the urgency level, severity is the frequency of occurrence and status is the current status of the issue such as committed or done. People play different roles as they interact with reports in ITS. The person who submits the report is the reporter and the assignee is the person to whom the issue is assigned. If the issue has been reported previously, historical changes are shown in the revision history. The estimated time is the time spent on the development and varies depending on the effort metric. The fields offered by each ITS vary, and not all fields are filled out for each project. Our strategy makes use of the fields that are largely filled in. Life Cycle of an Issue The series of phases and phase changes that an issue experiences throughout its life cycle is referred to as a “workflow.” Workflows for issues typically represent development cycles and business processes. Figure 2 shows a standard workflow of JIRA. The following stages of the JIRA workflow must be monitored as soon as an issue is created: – Open: The issue is open after creation and can be assigned to the assignee to begin working on it. – In Progress: The assignee has taken the initiative to begin working on the issue. – Resolved: The issue’s sub-tasks and works have all been finished. The reporter is currently waiting to confirm the matter. If verification is successful, the case will be closed or reopened depending on whether any additional changes are needed. – Reopened: Although this issue has already been solved, the solution is either flawed, omitted some important details, or needs some modifications. Issues are classified as assigned or resolved at the Reopened stage. – Closed: The issue is now regarded as resolved, and the current solution is accurate. Issues that have been closed can be reopened at a later time as needed. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. Authors: (1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr}; (2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Authors: Authors: (1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr}; (2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Table of Links Abstract and Introduction Background Approach Experiments and Results Discussion and Qualitative Analysis Related Work Conclusions and Future Work, and References Abstract and Introduction Abstract and Introduction Background Background Approach Approach Experiments and Results Experiments and Results Discussion and Qualitative Analysis Discussion and Qualitative Analysis Related Work Related Work Conclusions and Future Work, and References Conclusions and Future Work, and References 2 Background To understand our approach, it is essential to have a solid understanding of the structure of software issues in Issue Tracking Systems (ITS). 2.1 Software Issues in ITS This section explores software issues within ITS, covering their anatomy, life cycle, and significance in software projects. Issue Tracking Systems (ITS) An issue tracking system (ITS) is a software application that manages and maintains lists of issues. It provides a database of issue reports for a software project. Members of the software development team stay aligned and collaborate faster with the help of these tools. Users can log in with a password and post a new issue report or comment on an already-posted issue. There are various issue-tracking tools that are used in software development projects, such as JIRA, Bugzilla, and Azure DevOps. Although there are some differences among these tools, they all share a similar basic structure. We developed our approach with a data set of a company that uses JIRA software. Issue Tracking Systems (ITS) Anatomy of an Issue The issue report in ITS contains various fields, each of which contains a different piece of information as shown in Figure 1. The following details [48] could typically be included in the generic process of reporting an issue: Anatomy of an Issue – Summary: Title of the issue. – Description: Issue details including its cause, location, and timing. – Version: Version of the project to which the issue belongs. – Component: The predetermined component on which the issue depends. – Attachment: Attaches such as pictures, videos, and documents uploaded to the issue. – Priority: Urgency level of the issue. – Severity: Frequent occurrence and systemic effects. – Status: Current status of the issue. – Reporter: Person who creates the issue. – Assignee: Person to whom the issue is assigned. – Revision History: Historical changes of the issue. – Estimated time: Estimated time spent to develop or solve the issue. – Comments: Additional details that can be used to understand the issue. The summary, description, and comments all contain textual details about the issue. In our methodology, we extract numerical features using the language structure of summary and description. Version is the current version of the project which differs with each release. The issue can have a predefined tag or component added to it. (e.g. project v1.1.0) Users can upload files to help others understand the issue, an attachment refers to it. Priority, severity, and status are the categorical features of the issue. Priority is the urgency level, severity is the frequency of occurrence and status is the current status of the issue such as committed or done. People play different roles as they interact with reports in ITS. The person who submits the report is the reporter and the assignee is the person to whom the issue is assigned. If the issue has been reported previously, historical changes are shown in the revision history. The estimated time is the time spent on the development and varies depending on the effort metric. The fields offered by each ITS vary, and not all fields are filled out for each project. Our strategy makes use of the fields that are largely filled in. Life Cycle of an Issue The series of phases and phase changes that an issue experiences throughout its life cycle is referred to as a “workflow.” Workflows for issues typically represent development cycles and business processes. Figure 2 shows a standard workflow of JIRA. The following stages of the JIRA workflow must be monitored as soon as an issue is created: Life Cycle of an Issue – Open: The issue is open after creation and can be assigned to the assignee to begin working on it. – In Progress: The assignee has taken the initiative to begin working on the issue. – Resolved: The issue’s sub-tasks and works have all been finished. The reporter is currently waiting to confirm the matter. If verification is successful, the case will be closed or reopened depending on whether any additional changes are needed. – Reopened: Although this issue has already been solved, the solution is either flawed, omitted some important details, or needs some modifications. Issues are classified as assigned or resolved at the Reopened stage. – Closed: The issue is now regarded as resolved, and the current solution is accurate. Issues that have been closed can be reopened at a later time as needed. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. available on arxiv

Comparison of Machine Learning Methods: Background

Comparison of Machine Learning Methods: Abstract and Introduction

Authors:
(1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr};
(2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Table of Links Abstract and Introduction
Background
Approach
Experiments and Results
Discussion and Qualitative Analysis
Related Work
Conclusions and Future Work, and References 4 Experiments and Results This section presents the findings and outcomes of the conducted experiments, providing a comprehensive analysis of the collected data. We critically assess the performance and effectiveness of the proposed methodology by employing various evaluation metrics and techniques. We compare the extracted features from different perspectives, evaluating their individual contributions to the classification task. Lastly, we present a thorough statistical examination of the obtained results, employing appropriate statistical tests and measures to validate the significance and reliability of the findings. 4.1 Evaluation Table 5 presents the experiment results for the issue assignment. The models are evaluated on Team Assignment (TA) and Developer Assignment (DA). The stacking model (RF and Linear SVC) achieved the highest accuracy, with values of 0.92 for TA and 0.89 for DA. Other models, such as Support Vector Machine, Logistic Regression, and Random Forest, also showed good performance with accuracies ranging from 0.86 to 0.88. The transformer-based models, including DistilBert, Roberta, and Electra, demonstrated competitive accuracies, with Roberta and Electra achieving the highest scores in some cases. Table 6 provides the performance metrics for each class in the Stacking algorithm for issue assignment. Under the Team Assignment (TA) approach, the Stacking algorithm achieved a high precision value of 0.90 for the Developer class, indicating a low rate of false positive assignments. The Recall score of 0.99 for the Developer class demonstrates the algorithm’s ability to correctly identify the majority of instances assigned to developers. The Tester class shows a balance between precision (0.95) and recall (0.71), indicating accurate assignments with a relatively high rate of false negatives. The Designer class exhibits similar trends with a precision of 0.92 and a recall of 0.67. The Leader class has relatively lower precision and recall scores, indicating more challenging assignments for the algorithm. Under the Developer Assignment (DA) approach, the Stacking algorithm achieved high precision values for the Senior class (0.90) and the Mid class (0.89), indicating accurate assignments with low rates of false positives. The Mid class also demonstrates a high recall score of 0.94, indicating effective identification of instances assigned to this class. The Junior class shows a lower precision (0.62) and recall (0.53) compared to the other classes, suggesting potential challenges in accurately assigning instances to this class. We also test our classification algorithms using the most popular word embedding techniques to determine how well our features work. Figure 5 illustrates the comparison of our feature set with Tf-Idf and BOW methods for the issue assignment. Despite the potential of Word2Vec as a word embedding algorithm, the accuracy results in my approach do not yield comparable outcomes. We use the accuracy score as the comparison metric. The graph demonstrates that using our feature set yields superior results while using Tf-Idf and BOW yields comparable results. 4.2 Feature Comparison Reducing the number of redundant and irrelevant features is an effective way to improve the running time and generalization capability of a learning algorithm [12]. Feature selection methods are used to choose a subset of relevant features that contribute to the intended concept. These methods can employ a variety of models and techniques to calculate feature importance scores. One simple approach [9] is to calculate the coefficient statistics between each feature and the target variable. This method can help to identify the most important features of a given problem and discard the redundant or irrelevant ones. By reducing the number of features used for training a model, the running time of the algorithm can be significantly reduced without sacrificing accuracy. Moreover, feature selection can also improve the interpretability of the model, as it helps to identify the key factors that influence the target variable. We present the coefficient values of each feature in Figure 6. We find that Issue Type, namely FJC1, emerges as the most influential feature from our feature set. Apart from the Issue Type, we discover that the features Watchers and Summary Words also exhibit significant effectiveness in our analysis. Conversely, features such as Reopen Count, Test Execution Type, and Approval Type demonstrate no impact on our issue assignment process. In Figure 7, we present the effective POS tags in Turkish, highlighting the most influential ones among all POS tags. Notably, the number of unknown words, verbs, and nouns emerge as the most impactful features. Following the rigorous selection of the best features, we proceed to employ Scikit-learn’s [38] SelectFromModel, a powerful meta-transformer designed to choose features based on their importance weights, to retrain our models. Through this process, we carefully identify and select the top eight features that exhibited the highest significance, as determined by the module. Remarkably, leveraging this refined feature subset allows us to achieve optimal performance and attain the most favorable outcome in our experiments. 4.3 Statistical Analysis Statistical significance tests are conducted to compare classification methods for determining whether one learning algorithm outperforms another on a particular learning task. Dietterich [14] reviews five approximate statistical tests and concludes that McNemar’s test and the 5x2 cv t-test, both have low type I error and sufficient power. In our study, we combine all data sets into a single data set for the classification algorithm. Dietterich [14] recommends using a 5x2 ttest to statistically compare two classifiers on a single data set. The 5x2 f-test, which is also suggested as the new standard by the original authors above, is further expanded upon by Alpaydın [4]. The following lists the null hypothesis and the alternative hypothesis. The null hypothesis is that the probabilities are the same, or in simpler terms, neither of the two models outperforms the other. The alternative hypothesis, therefore, holds that the performances of the two models are not equivalent. Accordingly, we apply the 5x2 f-test implemented by Alpaydın [4] which is an extension of the 5x2 cv t-test as stated above. We create the matrix for all pairwise comparisons of learning algorithms. In this test, the splitting process (50% training and 50% test data) is repeated five times. A and B are fitted to the training split and their performance on the test split in each of the five iterations is assessed. The training and test sets are then rotated (the training set becomes the test set, and vice versa), and the performance is computed again, yielding two performance difference measures. Then, the mean and variance of the differences are estimated and the f-statistic proposed by Alpaydın is calculated as This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. Authors: (1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr}; (2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Authors: Authors: (1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr}; (2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Table of Links Abstract and Introduction Background Approach Experiments and Results Discussion and Qualitative Analysis Related Work Conclusions and Future Work, and References Abstract and Introduction Abstract and Introduction Background Background Approach Approach Experiments and Results Experiments and Results Discussion and Qualitative Analysis Discussion and Qualitative Analysis Related Work Related Work Conclusions and Future Work, and References Conclusions and Future Work, and References 4 Experiments and Results This section presents the findings and outcomes of the conducted experiments, providing a comprehensive analysis of the collected data. We critically assess the performance and effectiveness of the proposed methodology by employing various evaluation metrics and techniques. We compare the extracted features from different perspectives, evaluating their individual contributions to the classification task. Lastly, we present a thorough statistical examination of the obtained results, employing appropriate statistical tests and measures to validate the significance and reliability of the findings. 4.1 Evaluation Table 5 presents the experiment results for the issue assignment. The models are evaluated on Team Assignment (TA) and Developer Assignment (DA). The stacking model (RF and Linear SVC) achieved the highest accuracy, with values of 0.92 for TA and 0.89 for DA. Other models, such as Support Vector Machine, Logistic Regression, and Random Forest, also showed good performance with accuracies ranging from 0.86 to 0.88. The transformer-based models, including DistilBert, Roberta, and Electra, demonstrated competitive accuracies, with Roberta and Electra achieving the highest scores in some cases. Table 6 provides the performance metrics for each class in the Stacking algorithm for issue assignment. Under the Team Assignment (TA) approach, the Stacking algorithm achieved a high precision value of 0.90 for the Developer class, indicating a low rate of false positive assignments. The Recall score of 0.99 for the Developer class demonstrates the algorithm’s ability to correctly identify the majority of instances assigned to developers. The Tester class shows a balance between precision (0.95) and recall (0.71), indicating accurate assignments with a relatively high rate of false negatives. The Designer class exhibits similar trends with a precision of 0.92 and a recall of 0.67. The Leader class has relatively lower precision and recall scores, indicating more challenging assignments for the algorithm. Under the Developer Assignment (DA) approach, the Stacking algorithm achieved high precision values for the Senior class (0.90) and the Mid class (0.89), indicating accurate assignments with low rates of false positives. The Mid class also demonstrates a high recall score of 0.94, indicating effective identification of instances assigned to this class. The Junior class shows a lower precision (0.62) and recall (0.53) compared to the other classes, suggesting potential challenges in accurately assigning instances to this class. We also test our classification algorithms using the most popular word embedding techniques to determine how well our features work. Figure 5 illustrates the comparison of our feature set with Tf-Idf and BOW methods for the issue assignment. Despite the potential of Word2Vec as a word embedding algorithm, the accuracy results in my approach do not yield comparable outcomes. We use the accuracy score as the comparison metric. The graph demonstrates that using our feature set yields superior results while using Tf-Idf and BOW yields comparable results. 4.2 Feature Comparison Reducing the number of redundant and irrelevant features is an effective way to improve the running time and generalization capability of a learning algorithm [12]. Feature selection methods are used to choose a subset of relevant features that contribute to the intended concept. These methods can employ a variety of models and techniques to calculate feature importance scores. One simple approach [9] is to calculate the coefficient statistics between each feature and the target variable. This method can help to identify the most important features of a given problem and discard the redundant or irrelevant ones. By reducing the number of features used for training a model, the running time of the algorithm can be significantly reduced without sacrificing accuracy. Moreover, feature selection can also improve the interpretability of the model, as it helps to identify the key factors that influence the target variable. We present the coefficient values of each feature in Figure 6. We find that Issue Type, namely FJC1, emerges as the most influential feature from our feature set. Apart from the Issue Type, we discover that the features Watchers and Summary Words also exhibit significant effectiveness in our analysis. Conversely, features such as Reopen Count, Test Execution Type, and Approval Type demonstrate no impact on our issue assignment process. In Figure 7, we present the effective POS tags in Turkish, highlighting the most influential ones among all POS tags. Notably, the number of unknown words, verbs, and nouns emerge as the most impactful features. Following the rigorous selection of the best features, we proceed to employ Scikit-learn’s [38] SelectFromModel, a powerful meta-transformer designed to choose features based on their importance weights, to retrain our models. Through this process, we carefully identify and select the top eight features that exhibited the highest significance, as determined by the module. Remarkably, leveraging this refined feature subset allows us to achieve optimal performance and attain the most favorable outcome in our experiments. 4.3 Statistical Analysis Statistical significance tests are conducted to compare classification methods for determining whether one learning algorithm outperforms another on a particular learning task. Dietterich [14] reviews five approximate statistical tests and concludes that McNemar’s test and the 5x2 cv t-test, both have low type I error and sufficient power. In our study, we combine all data sets into a single data set for the classification algorithm. Dietterich [14] recommends using a 5x2 ttest to statistically compare two classifiers on a single data set. The 5x2 f-test, which is also suggested as the new standard by the original authors above, is further expanded upon by Alpaydın [4]. The following lists the null hypothesis and the alternative hypothesis. The null hypothesis is that the probabilities are the same, or in simpler terms, neither of the two models outperforms the other. The alternative hypothesis, therefore, holds that the performances of the two models are not equivalent. Accordingly, we apply the 5x2 f-test implemented by Alpaydın [4] which is an extension of the 5x2 cv t-test as stated above. We create the matrix for all pairwise comparisons of learning algorithms. In this test, the splitting process (50% training and 50% test data) is repeated five times. A and B are fitted to the training split and their performance on the test split in each of the five iterations is assessed. The training and test sets are then rotated (the training set becomes the test set, and vice versa), and the performance is computed again, yielding two performance difference measures. Then, the mean and variance of the differences are estimated and the f-statistic proposed by Alpaydın is calculated as This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. available on arxiv

Comparison of Machine Learning Methods: Experiments and Results

Authors:
(1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr};
(2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Table of Links Abstract and Introduction
Background
Approach
Experiments and Results
Discussion and Qualitative Analysis
Related Work
Conclusions and Future Work, and References 6 Related Work Several studies in the literature have focused on issue classification, which has addressed a variety of objectives, including issue assignment, effort estimation, issue prioritization, and so on. In this section, we briefly give details regarding issue assignment studies in general and all Turkish-language issue classification studies in particular. Several types of research have been conducted in order to automate the time-consuming task of issue assignment. In 2017, Goyal et al. [18] review and categorize 75 research papers on the automated bug assignment area. They identify seven categories: machine learning [25,47,55,5,8,36], information retrieval [58], auction [24], social network [57], tossing graph [50], fuzzy set [37] and operational research based [26] techniques. They capture the fact that for automatic bug report assignment, machine learning and information retrieval techniques are the most popular ones. In recent years, deep learning algorithms have also been successfully applied in this field, which has recently revolutionized the idea of word sequence representation and demonstrated encouraging advancements in a number of classification tasks [17]. In this section, we restrict our focus to machine learning and deep learning architectures used to train issue assignment systems. The machine learning algorithms use historical bug reports to build a supervised or unsupervised machine learning classifier, which is then used to choose appropriate developers for new bug reports. Naive Bayes is the most widely used classifier in machine learning-based approaches according to prior studies [56,8,6,30,47], and it has been extensively tested [18] in the bug reports of open-source projects. Most studies use Eclipse [8,5,3,51,2,39] and Mozilla [8,51,39,23] projects to validate their proposals. Machine learning models in most approaches [5,36,51] use only summary and description as textual features of the issues. Jonsson et al. [25] use the combined title and description as textual features and version, type, priority, and submitter columns as nominal features. Sharma et al. [46] consider bug attributes, namely, severity, priority, component, operating system, and the bug assignee. To estimate the value of terms, most of the approaches [51,25,47,42] in the literature employ term-weighting techniques like Tf-Idf. Jonsson et al. [25] represent textual parts in the bug reports as the 100 words with the highest Tf-Idf. Shokripour et al. [47] use time metadata in Tf-Idf (Time-Tf-Idf). To determine the value of terms in a document and corpus, the Tf-Idf technique only considers their frequency. However, in determining the weight, time-based Tf-Idf considers the time spent using the term in the project. The developer’s recent use of the term is taken into account when determining the value of the developer’s expertise. They rank the developers according to their calculated term expertise, and the first developer on the list is assigned to fix the new bug. However, prior studies focused on open-source projects only but rarely [25,36] attempted in industrial environments like our study. Jonsson et al. [25] use ensemble learner Stacked Generalization, which is our best method also, that combines several machine learning classifiers on data from the automation and telecommunication company. In their approach, the different classes correspond to the development teams. Oliveira et al. [36] also use the data set of a large electronic company. They create a model that can distribute new issues according to the responsibilities of the teams using a variety of machine learning techniques and the WEKA [21] tool. To improve prediction accuracy, some researchers use incremental learning methods. Bhattacharya et al. [8] use various machine learning algorithms and achieve the best results using the NB classifier in combination with the product-component features, tossing graphs, and incremental learning in mostly used large projects: Mozilla and Eclipse. Xia et al. [55] offer the multi-feature topic model (MTM), a specialized topic modeling approach that extends Latent Dirichlet Allocation (LDA) for the bug assignment. To map the term space to the subject space, their approach takes into account product and component information from issue reports. Then, they suggest an incremental learning mechanism that uses the topic distribution of a new bug report to assign an appropriate developer based on the reports that the developer has previously fixed. The deep learning algorithms are attempted first in 2017 [16] for bug report assignment recommendation, to the best of our knowledge. Gupta et al. [19] describe the popular deep learning approaches applied to the domain of bug reports and Recurrent Neural Networks (RNN) and Long Short Term Memory (LSTM) are a few famous approaches being used for the deep learning-based approaches [16,31]. Mani et al. [31] use title and description parts and Florea et al. [16] use the component id, product id, and bug severity fields as one-hot-encoded categorical variables in addition to title, description, and comments to represent the issues. In 2022, Feng et al. [15] use four transformers models BERT and RoBERTa along with their distilled counterparts DistilBERT, DistilRoBERTa in an ensemble using a resolver team, resolver person, and description columns of the issues. In research using Turkish issue reports, there are limited studies available in a few fields. The reason may be the agglutinative nature of the Turkish language and the absence of a shared data set for Turkish issues in the literature. Aktas et al. [1] classified the issues they gathered from the banking industry among various software development teams. They use the Jira issue reports for their research like our study. They use SVC, CNN, and LSTM models to solve the classification problem, and they represent the summary and description columns of the issue reports as ordered vectors of Tf-Idf scores. The linear SVC classifier offers the best assignment accuracy for their research with a 0.82 score. Koksal et al. [27] present an automated bug classification approach using a commercial proprietary bug data set. They apply several machine learning algorithms and the SVM classifier is the best algorithm with 0.72 accuracy. In 2022, Tunali [52] prioritizes the software development demands of a private insurance company in Turkey. He proposes several deep-learning architectures and a pre-trained transformer model called distilbert-base-turkish-cased based on DistilBERT to achieve the highest accuracy of 0.82. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. Authors: (1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr}; (2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Authors: Authors: (1) Busra Tabak [0000 −0001 −7460 −3689], Bogazici University, Turkey {busra.tabak@boun.edu.tr}; (2) Fatma Basak Aydemir [0000 −0003 −3833 −3997],  Bogazici University, Turkey {basak.aydemir@boun.edu.tr}. Table of Links Abstract and Introduction Background Approach Experiments and Results Discussion and Qualitative Analysis Related Work Conclusions and Future Work, and References Abstract and Introduction Abstract and Introduction Background Background Approach Approach Experiments and Results Experiments and Results Discussion and Qualitative Analysis Discussion and Qualitative Analysis Related Work Related Work Conclusions and Future Work, and References Conclusions and Future Work, and References 6 Related Work Several studies in the literature have focused on issue classification, which has addressed a variety of objectives, including issue assignment, effort estimation, issue prioritization, and so on. In this section, we briefly give details regarding issue assignment studies in general and all Turkish-language issue classification studies in particular. Several types of research have been conducted in order to automate the time-consuming task of issue assignment. In 2017, Goyal et al. [18] review and categorize 75 research papers on the automated bug assignment area. They identify seven categories: machine learning [25,47,55,5,8,36], information retrieval [58], auction [24], social network [57], tossing graph [50], fuzzy set [37] and operational research based [26] techniques. They capture the fact that for automatic bug report assignment, machine learning and information retrieval techniques are the most popular ones. In recent years, deep learning algorithms have also been successfully applied in this field, which has recently revolutionized the idea of word sequence representation and demonstrated encouraging advancements in a number of classification tasks [17]. In this section, we restrict our focus to machine learning and deep learning architectures used to train issue assignment systems. The machine learning algorithms use historical bug reports to build a supervised or unsupervised machine learning classifier, which is then used to choose appropriate developers for new bug reports. Naive Bayes is the most widely used classifier in machine learning-based approaches according to prior studies [56,8,6,30,47], and it has been extensively tested [18] in the bug reports of open-source projects. Most studies use Eclipse [8,5,3,51,2,39] and Mozilla [8,51,39,23] projects to validate their proposals. Machine learning models in most approaches [5,36,51] use only summary and description as textual features of the issues. Jonsson et al. [25] use the combined title and description as textual features and version, type, priority, and submitter columns as nominal features. Sharma et al. [46] consider bug attributes, namely, severity, priority, component, operating system, and the bug assignee. To estimate the value of terms, most of the approaches [51,25,47,42] in the literature employ term-weighting techniques like Tf-Idf. Jonsson et al. [25] represent textual parts in the bug reports as the 100 words with the highest Tf-Idf. Shokripour et al. [47] use time metadata in Tf-Idf (Time-Tf-Idf). To determine the value of terms in a document and corpus, the Tf-Idf technique only considers their frequency. However, in determining the weight, time-based Tf-Idf considers the time spent using the term in the project. The developer’s recent use of the term is taken into account when determining the value of the developer’s expertise. They rank the developers according to their calculated term expertise, and the first developer on the list is assigned to fix the new bug. However, prior studies focused on open-source projects only but rarely [25,36] attempted in industrial environments like our study. Jonsson et al. [25] use ensemble learner Stacked Generalization, which is our best method also, that combines several machine learning classifiers on data from the automation and telecommunication company. In their approach, the different classes correspond to the development teams. Oliveira et al. [36] also use the data set of a large electronic company. They create a model that can distribute new issues according to the responsibilities of the teams using a variety of machine learning techniques and the WEKA [21] tool. To improve prediction accuracy, some researchers use incremental learning methods. Bhattacharya et al. [8] use various machine learning algorithms and achieve the best results using the NB classifier in combination with the product-component features, tossing graphs, and incremental learning in mostly used large projects: Mozilla and Eclipse. Xia et al. [55] offer the multi-feature topic model (MTM), a specialized topic modeling approach that extends Latent Dirichlet Allocation (LDA) for the bug assignment. To map the term space to the subject space, their approach takes into account product and component information from issue reports. Then, they suggest an incremental learning mechanism that uses the topic distribution of a new bug report to assign an appropriate developer based on the reports that the developer has previously fixed. The deep learning algorithms are attempted first in 2017 [16] for bug report assignment recommendation, to the best of our knowledge. Gupta et al. [19] describe the popular deep learning approaches applied to the domain of bug reports and Recurrent Neural Networks (RNN) and Long Short Term Memory (LSTM) are a few famous approaches being used for the deep learning-based approaches [16,31]. Mani et al. [31] use title and description parts and Florea et al. [16] use the component id, product id, and bug severity fields as one-hot-encoded categorical variables in addition to title, description, and comments to represent the issues. In 2022, Feng et al. [15] use four transformers models BERT and RoBERTa along with their distilled counterparts DistilBERT, DistilRoBERTa in an ensemble using a resolver team, resolver person, and description columns of the issues. In research using Turkish issue reports, there are limited studies available in a few fields. The reason may be the agglutinative nature of the Turkish language and the absence of a shared data set for Turkish issues in the literature. Aktas et al. [1] classified the issues they gathered from the banking industry among various software development teams. They use the Jira issue reports for their research like our study. They use SVC, CNN, and LSTM models to solve the classification problem, and they represent the summary and description columns of the issue reports as ordered vectors of Tf-Idf scores. The linear SVC classifier offers the best assignment accuracy for their research with a 0.82 score. Koksal et al. [27] present an automated bug classification approach using a commercial proprietary bug data set. They apply several machine learning algorithms and the SVM classifier is the best algorithm with 0.72 accuracy. In 2022, Tunali [52] prioritizes the software development demands of a private insurance company in Turkey. He proposes several deep-learning architectures and a pre-trained transformer model called distilbert-base-turkish-cased based on DistilBERT to achieve the highest accuracy of 0.82. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. available on arxiv

FATMAP

fatmap.com

51-200 employees

Since 2013

COMPANY RANKING

FATMAP

EVERGREEN INDEX #4256

Representing this company? Click here to claim and customize this page!

FATMAP WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

FATMAP

fatmap.com

51-200 employees

Since 2013

COMPANY RANKING

FATMAP

EVERGREEN INDEX #4256

RELATED COMPANIESRANK

STORIES ABOUT FATMAP

Representing this company? Click here to claim and customize this page!

FATMAP WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!